register vitpose-b/l/h for coco

bug fix

update readme

rename

add giant results

sync drive path

trest

CITATION.cff

@@ -0,0 +1,8 @@
+cff-version: 1.2.0
+message: "If you use this software, please cite it as below."
+authors:
+  - name: "MMPose Contributors"
+title: "OpenMMLab Pose Estimation Toolbox and Benchmark"
+date-released: 2020-08-31
+url: "https://github.com/open-mmlab/mmpose"
+license: Apache-2.0

LICENSE

@@ -0,0 +1,203 @@
+Copyright 2018-2020 Open-MMLab. All rights reserved.
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright 2018-2020 Open-MMLab.
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

MANIFEST.in

@@ -0,0 +1,5 @@
+include requirements/*.txt
+include mmpose/.mim/model-index.yml
+recursive-include mmpose/.mim/configs *.py *.yml
+recursive-include mmpose/.mim/tools *.py *.sh
+recursive-include mmpose/.mim/demo *.py

README.md

@@ -19,15 +19,80 @@
 
 This branch contains the pytorch implementation of <a href="https://arxiv.org/abs/2204.12484">ViTPose: Simple Vision Transformer Baselines for Human Pose Estimation</a>. It obtains 81.1 AP on MS COCO Keypoint test-dev set.
 
-## Results from this repo on MS COCO val set
+## Results from this repo on MS COCO val set (single task training)
 
-Using detection results from a detector that obtains 56 mAP on person.
+Using detection results from a detector that obtains 56 mAP on person. The configs here are for both training and test.
+
+> With classic decoder
 
 | Model | Pretrain | Resolution | AP | AR | config | log | weight |
 | :----: | :----: | :----: | :----: | :----: | :----: | :----: | :----: |
-| ViT-Base | MAE | 256x192 | 75.8 | 81.1 | config | [log](logs/vit-base.log.json) |  |
+| ViTPose-B | MAE | 256x192 | 75.8 | 81.1 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_base_coco_256x192.py) | [log](logs/vitpose-b.log.json) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSMjp1_NrV3VRSmK?e=Q1uZKs) |
-| ViT-Large | MAE | 256x192 | 78.3 | 83.5 | config | [log](logs/vit-large.log.json) |  |
+| ViTPose-L | MAE | 256x192 | 78.3 | 83.5 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_large_coco_256x192.py) | [log](logs/vitpose-l.log.json) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSd9k_kuktPtiP4F?e=K7DGYT) |
-| ViT-Huge | MAE | 256x192 | 79.1 | 84.1 | config | [log](logs/vit-huge.log.json) |  |
+| ViTPose-H | MAE | 256x192 | 79.1 | 84.1 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_huge_coco_256x192.py) | [log](logs/vitpose-h.log.json) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgShLMI-kkmvNfF_h?e=dEhGHe) |
+
+> With simple decoder
+
+| Model | Pretrain | Resolution | AP | AR | config | log | weight |
+| :----: | :----: | :----: | :----: | :----: | :----: | :----: | :----: |
+| ViTPose-B | MAE | 256x192 | 75.5 | 80.9 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_base_simple_coco_256x192.py) | [log](logs/vitpose-b-simple.log.json) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSRPKrD5PmDRiv0R?e=jifvOe) |
+| ViTPose-L | MAE | 256x192 | 78.2 | 83.4 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_large_simple_coco_256x192.py) | [log](logs/vitpose-l-simple.log.json) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSVS6DP2LmKwZ3sm?e=MmCvDT) |
+| ViTPose-H | MAE | 256x192 | 78.9 | 84.0 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_huge_simple_coco_256x192.py) | [log](logs/vitpose-h-simple.log.json) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSbHyN2mjh2n2LyG?e=y0FgMK) |
+
+## Results from this repo on MS COCO val set (multi task training)
+
+Using detection results from a detector that obtains 56 mAP on person. Note the configs here are only for evaluation.
+
+| Model | Dataset | Resolution | AP | AR | config | weight |
+| :----: | :----: | :----: | :----: | :----: | :----: | :----: |
+| ViTPose-B | COCO+AIC+MPII+CrowdPose | 256x192 | 77.5 | 82.6 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_base_coco_256x192.py)  |[Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSrlMB093JzJtqq-?e=Jr5S3R) |
+| ViTPose-L | COCO+AIC+MPII+CrowdPose | 256x192 | 79.1 | 84.1 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_large_coco_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgTBm3dCVmBUbHYT6?e=fHUrTq) |
+| ViTPose-H | COCO+AIC+MPII+CrowdPose | 256x192 | 79.8 | 84.8 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/coco/ViTPose_huge_coco_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgS5rLeRAJiWobCdh?e=41GsDd) |
+| ViTPose-G | COCO+AIC+MPII+CrowdPose | 576x432 | 81.0 | 85.6 | | |
+
+## Results from this repo on OCHuman test set (multi task training)
+
+Using groundtruth bounding boxes. Note the configs here are only for evaluation.
+
+| Model | Dataset | Resolution | AP | AR | config | weight |
+| :----: | :----: | :----: | :----: | :----: | :----: | :----: |
+| ViTPose-B | COCO+AIC+MPII+CrowdPose | 256x192 | 88.2 | 90.0 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/ochuman/ViTPose_base_ochuman_256x192.py)  |[Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSrlMB093JzJtqq-?e=Jr5S3R) |
+| ViTPose-L | COCO+AIC+MPII+CrowdPose | 256x192 | 91.5 | 92.8 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/ochuman/ViTPose_large_ochuman_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgTBm3dCVmBUbHYT6?e=fHUrTq) |
+| ViTPose-H | COCO+AIC+MPII+CrowdPose | 256x192 | 91.6 | 92.8 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/ochuman/ViTPose_huge_ochuman_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgS5rLeRAJiWobCdh?e=41GsDd) |
+| ViTPose-G | COCO+AIC+MPII+CrowdPose | 576x432 | 93.3 | 94.3 | | |
+
+## Results from this repo on CrowdPose test set (multi task training)
+
+Using YOLOv3 human detector. Note the configs here are only for evaluation.
+
+| Model | Dataset | Resolution | AP | AP(H) | config | weight |
+| :----: | :----: | :----: | :----: | :----: | :----: | :----: |
+| ViTPose-B | COCO+AIC+MPII+CrowdPose | 256x192 | 74.7 | 63.3 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/crowdpose/ViTPose_base_crowdpose_256x192.py)  |[Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgStrrCb91cPlaxJx?e=6Xobo6) |
+| ViTPose-L | COCO+AIC+MPII+CrowdPose | 256x192 | 76.6 | 65.9 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/crowdpose/ViTPose_large_crowdpose_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgTK3dug-r7c6GFyu?e=1ZBpEG) |
+| ViTPose-H | COCO+AIC+MPII+CrowdPose | 256x192 | 76.3 | 65.6 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/crowdpose/ViTPose_huge_crowdpose_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgS-oAvEV4MTD--Xr?e=EeW2Fu) |
+| ViTPose-G | COCO+AIC+MPII+CrowdPose | 576x432 | 78.3 | 67.9 | | |
+
+## Results from this repo on MPII val set (multi task training)
+
+Using groundtruth bounding boxes. Note the configs here are only for evaluation. The metric is PCKh.
+
+| Model | Dataset | Resolution | Mean | config | weight |
+| :----: | :----: | :----: | :----: | :----: | :----: |
+| ViTPose-B | COCO+AIC+MPII+CrowdPose | 256x192 | 93.4 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/mpii/ViTPose_base_mpii_256x192.py)  |[Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSy_OSEm906wd2LB?e=GOSg14) |
+| ViTPose-L | COCO+AIC+MPII+CrowdPose | 256x192 | 93.9 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/mpii/ViTPose_large_mpii_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgTM32I6Kpjr-esl6?e=qvh0Yl) |
+| ViTPose-H | COCO+AIC+MPII+CrowdPose | 256x192 | 94.1 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/mpii/ViTPose_huge_mpii_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgTT90XEQBKy-scIH?e=D2WhTS) |
+| ViTPose-G | COCO+AIC+MPII+CrowdPose | 576x432 | 94.3 | | |
+
+## Results from this repo on AI Challenger test set (multi task training)
+
+Using groundtruth bounding boxes. Note the configs here are only for evaluation.
+
+| Model | Dataset | Resolution | AP | AR | config | weight |
+| :----: | :----: | :----: | :----: | :----: | :----: | :----: |
+| ViTPose-B | COCO+AIC+MPII+CrowdPose | 256x192 | 31.9 | 36.3 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/aic/ViTPose_base_aic_256x192.py)  |[Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgSlvdVaXTC92SHYH?e=j7iqcp) |
+| ViTPose-L | COCO+AIC+MPII+CrowdPose | 256x192 | 34.6 | 39.0 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/aic/ViTPose_base_aic_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgTF06FX3FSAm0MOH?e=rYts9F) |
+| ViTPose-H | COCO+AIC+MPII+CrowdPose | 256x192 | 35.3 | 39.8 | [config](configs/body/2d_kpt_sview_rgb_img/topdown_heatmap/aic/ViTPose_base_aic_256x192.py) | [Onedrive](https://1drv.ms/u/s!AimBgYV7JjTlgS1MRmb2mcow_K04?e=q9jPab) |
+| ViTPose-G | COCO+AIC+MPII+CrowdPose | 576x432 | 43.2 | 47.1 | | |
 
 ## Updates
 
@@ -66,13 +131,21 @@ bash tools/dist_train.sh <Config PATH> <NUM GPUs> --cfg-options model.pretrained
 python -m torch.distributed.launch --nnodes <Num Machines> --node_rank <Rank of Machine> --nproc_per_node <GPUs Per Machine> --master_addr <Master Addr> --master_port <Master Port> tools/train.py <Config PATH> --cfg-options model.pretrained=<Pretrained PATH> --launcher pytorch --seed 0
 ```
 
+To test the pretrained models performance, please run 
+
+```bash
+bash tools/dist_test.sh <Config PATH> <Checkpoint PATH> <NUM GPUs>
+```
+
 ## Todo
 
 This repo current contains modifications including:
 
-- [ ] Upload configs and pretrained models
+- [x] Upload configs and pretrained models
+
+- [x] More models with SOTA results
 
-- [ ] More models with SOTA results
+- [ ] Upload multi-task training config
 
 ## Acknowledge
 We acknowledge the excellent implementation from [mmpose](https://github.com/open-mmlab/mmdetection) and [MAE](https://github.com/facebookresearch/mae).

configs/_base_/datasets/300w.py

@@ -0,0 +1,384 @@
+dataset_info = dict(
+    dataset_name='300w',
+    paper_info=dict(
+        author='Sagonas, Christos and Antonakos, Epameinondas '
+        'and Tzimiropoulos, Georgios and Zafeiriou, Stefanos '
+        'and Pantic, Maja',
+        title='300 faces in-the-wild challenge: '
+        'Database and results',
+        container='Image and vision computing',
+        year='2016',
+        homepage='https://ibug.doc.ic.ac.uk/resources/300-W/',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='kpt-0', id=0, color=[255, 255, 255], type='', swap='kpt-16'),
+        1:
+        dict(
+            name='kpt-1', id=1, color=[255, 255, 255], type='', swap='kpt-15'),
+        2:
+        dict(
+            name='kpt-2', id=2, color=[255, 255, 255], type='', swap='kpt-14'),
+        3:
+        dict(
+            name='kpt-3', id=3, color=[255, 255, 255], type='', swap='kpt-13'),
+        4:
+        dict(
+            name='kpt-4', id=4, color=[255, 255, 255], type='', swap='kpt-12'),
+        5:
+        dict(
+            name='kpt-5', id=5, color=[255, 255, 255], type='', swap='kpt-11'),
+        6:
+        dict(
+            name='kpt-6', id=6, color=[255, 255, 255], type='', swap='kpt-10'),
+        7:
+        dict(name='kpt-7', id=7, color=[255, 255, 255], type='', swap='kpt-9'),
+        8:
+        dict(name='kpt-8', id=8, color=[255, 255, 255], type='', swap=''),
+        9:
+        dict(name='kpt-9', id=9, color=[255, 255, 255], type='', swap='kpt-7'),
+        10:
+        dict(
+            name='kpt-10', id=10, color=[255, 255, 255], type='',
+            swap='kpt-6'),
+        11:
+        dict(
+            name='kpt-11', id=11, color=[255, 255, 255], type='',
+            swap='kpt-5'),
+        12:
+        dict(
+            name='kpt-12', id=12, color=[255, 255, 255], type='',
+            swap='kpt-4'),
+        13:
+        dict(
+            name='kpt-13', id=13, color=[255, 255, 255], type='',
+            swap='kpt-3'),
+        14:
+        dict(
+            name='kpt-14', id=14, color=[255, 255, 255], type='',
+            swap='kpt-2'),
+        15:
+        dict(
+            name='kpt-15', id=15, color=[255, 255, 255], type='',
+            swap='kpt-1'),
+        16:
+        dict(
+            name='kpt-16', id=16, color=[255, 255, 255], type='',
+            swap='kpt-0'),
+        17:
+        dict(
+            name='kpt-17',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-26'),
+        18:
+        dict(
+            name='kpt-18',
+            id=18,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-25'),
+        19:
+        dict(
+            name='kpt-19',
+            id=19,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-24'),
+        20:
+        dict(
+            name='kpt-20',
+            id=20,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-23'),
+        21:
+        dict(
+            name='kpt-21',
+            id=21,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-22'),
+        22:
+        dict(
+            name='kpt-22',
+            id=22,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-21'),
+        23:
+        dict(
+            name='kpt-23',
+            id=23,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-20'),
+        24:
+        dict(
+            name='kpt-24',
+            id=24,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-19'),
+        25:
+        dict(
+            name='kpt-25',
+            id=25,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-18'),
+        26:
+        dict(
+            name='kpt-26',
+            id=26,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-17'),
+        27:
+        dict(name='kpt-27', id=27, color=[255, 255, 255], type='', swap=''),
+        28:
+        dict(name='kpt-28', id=28, color=[255, 255, 255], type='', swap=''),
+        29:
+        dict(name='kpt-29', id=29, color=[255, 255, 255], type='', swap=''),
+        30:
+        dict(name='kpt-30', id=30, color=[255, 255, 255], type='', swap=''),
+        31:
+        dict(
+            name='kpt-31',
+            id=31,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-35'),
+        32:
+        dict(
+            name='kpt-32',
+            id=32,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-34'),
+        33:
+        dict(name='kpt-33', id=33, color=[255, 255, 255], type='', swap=''),
+        34:
+        dict(
+            name='kpt-34',
+            id=34,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-32'),
+        35:
+        dict(
+            name='kpt-35',
+            id=35,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-31'),
+        36:
+        dict(
+            name='kpt-36',
+            id=36,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-45'),
+        37:
+        dict(
+            name='kpt-37',
+            id=37,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-44'),
+        38:
+        dict(
+            name='kpt-38',
+            id=38,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-43'),
+        39:
+        dict(
+            name='kpt-39',
+            id=39,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-42'),
+        40:
+        dict(
+            name='kpt-40',
+            id=40,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-47'),
+        41:
+        dict(
+            name='kpt-41',
+            id=41,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-46'),
+        42:
+        dict(
+            name='kpt-42',
+            id=42,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-39'),
+        43:
+        dict(
+            name='kpt-43',
+            id=43,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-38'),
+        44:
+        dict(
+            name='kpt-44',
+            id=44,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-37'),
+        45:
+        dict(
+            name='kpt-45',
+            id=45,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-36'),
+        46:
+        dict(
+            name='kpt-46',
+            id=46,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-41'),
+        47:
+        dict(
+            name='kpt-47',
+            id=47,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-40'),
+        48:
+        dict(
+            name='kpt-48',
+            id=48,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-54'),
+        49:
+        dict(
+            name='kpt-49',
+            id=49,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-53'),
+        50:
+        dict(
+            name='kpt-50',
+            id=50,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-52'),
+        51:
+        dict(name='kpt-51', id=51, color=[255, 255, 255], type='', swap=''),
+        52:
+        dict(
+            name='kpt-52',
+            id=52,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-50'),
+        53:
+        dict(
+            name='kpt-53',
+            id=53,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-49'),
+        54:
+        dict(
+            name='kpt-54',
+            id=54,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-48'),
+        55:
+        dict(
+            name='kpt-55',
+            id=55,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-59'),
+        56:
+        dict(
+            name='kpt-56',
+            id=56,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-58'),
+        57:
+        dict(name='kpt-57', id=57, color=[255, 255, 255], type='', swap=''),
+        58:
+        dict(
+            name='kpt-58',
+            id=58,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-56'),
+        59:
+        dict(
+            name='kpt-59',
+            id=59,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-55'),
+        60:
+        dict(
+            name='kpt-60',
+            id=60,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-64'),
+        61:
+        dict(
+            name='kpt-61',
+            id=61,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-63'),
+        62:
+        dict(name='kpt-62', id=62, color=[255, 255, 255], type='', swap=''),
+        63:
+        dict(
+            name='kpt-63',
+            id=63,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-61'),
+        64:
+        dict(
+            name='kpt-64',
+            id=64,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-60'),
+        65:
+        dict(
+            name='kpt-65',
+            id=65,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-67'),
+        66:
+        dict(name='kpt-66', id=66, color=[255, 255, 255], type='', swap=''),
+        67:
+        dict(
+            name='kpt-67',
+            id=67,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-65'),
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 68,
+    sigmas=[])

configs/_base_/datasets/aflw.py

@@ -0,0 +1,83 @@
+dataset_info = dict(
+    dataset_name='aflw',
+    paper_info=dict(
+        author='Koestinger, Martin and Wohlhart, Paul and '
+        'Roth, Peter M and Bischof, Horst',
+        title='Annotated facial landmarks in the wild: '
+        'A large-scale, real-world database for facial '
+        'landmark localization',
+        container='2011 IEEE international conference on computer '
+        'vision workshops (ICCV workshops)',
+        year='2011',
+        homepage='https://www.tugraz.at/institute/icg/research/'
+        'team-bischof/lrs/downloads/aflw/',
+    ),
+    keypoint_info={
+        0:
+        dict(name='kpt-0', id=0, color=[255, 255, 255], type='', swap='kpt-5'),
+        1:
+        dict(name='kpt-1', id=1, color=[255, 255, 255], type='', swap='kpt-4'),
+        2:
+        dict(name='kpt-2', id=2, color=[255, 255, 255], type='', swap='kpt-3'),
+        3:
+        dict(name='kpt-3', id=3, color=[255, 255, 255], type='', swap='kpt-2'),
+        4:
+        dict(name='kpt-4', id=4, color=[255, 255, 255], type='', swap='kpt-1'),
+        5:
+        dict(name='kpt-5', id=5, color=[255, 255, 255], type='', swap='kpt-0'),
+        6:
+        dict(
+            name='kpt-6', id=6, color=[255, 255, 255], type='', swap='kpt-11'),
+        7:
+        dict(
+            name='kpt-7', id=7, color=[255, 255, 255], type='', swap='kpt-10'),
+        8:
+        dict(name='kpt-8', id=8, color=[255, 255, 255], type='', swap='kpt-9'),
+        9:
+        dict(name='kpt-9', id=9, color=[255, 255, 255], type='', swap='kpt-8'),
+        10:
+        dict(
+            name='kpt-10', id=10, color=[255, 255, 255], type='',
+            swap='kpt-7'),
+        11:
+        dict(
+            name='kpt-11', id=11, color=[255, 255, 255], type='',
+            swap='kpt-6'),
+        12:
+        dict(
+            name='kpt-12',
+            id=12,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-14'),
+        13:
+        dict(name='kpt-13', id=13, color=[255, 255, 255], type='', swap=''),
+        14:
+        dict(
+            name='kpt-14',
+            id=14,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-12'),
+        15:
+        dict(
+            name='kpt-15',
+            id=15,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-17'),
+        16:
+        dict(name='kpt-16', id=16, color=[255, 255, 255], type='', swap=''),
+        17:
+        dict(
+            name='kpt-17',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-15'),
+        18:
+        dict(name='kpt-18', id=18, color=[255, 255, 255], type='', swap='')
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 19,
+    sigmas=[])

configs/_base_/datasets/aic.py

@@ -0,0 +1,140 @@
+dataset_info = dict(
+    dataset_name='aic',
+    paper_info=dict(
+        author='Wu, Jiahong and Zheng, He and Zhao, Bo and '
+        'Li, Yixin and Yan, Baoming and Liang, Rui and '
+        'Wang, Wenjia and Zhou, Shipei and Lin, Guosen and '
+        'Fu, Yanwei and others',
+        title='Ai challenger: A large-scale dataset for going '
+        'deeper in image understanding',
+        container='arXiv',
+        year='2017',
+        homepage='https://github.com/AIChallenger/AI_Challenger_2017',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='right_shoulder',
+            id=0,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        1:
+        dict(
+            name='right_elbow',
+            id=1,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        2:
+        dict(
+            name='right_wrist',
+            id=2,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        3:
+        dict(
+            name='left_shoulder',
+            id=3,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        4:
+        dict(
+            name='left_elbow',
+            id=4,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        5:
+        dict(
+            name='left_wrist',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        6:
+        dict(
+            name='right_hip',
+            id=6,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        7:
+        dict(
+            name='right_knee',
+            id=7,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        8:
+        dict(
+            name='right_ankle',
+            id=8,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle'),
+        9:
+        dict(
+            name='left_hip',
+            id=9,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        10:
+        dict(
+            name='left_knee',
+            id=10,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        11:
+        dict(
+            name='left_ankle',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        12:
+        dict(
+            name='head_top',
+            id=12,
+            color=[51, 153, 255],
+            type='upper',
+            swap=''),
+        13:
+        dict(name='neck', id=13, color=[51, 153, 255], type='upper', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('right_wrist', 'right_elbow'), id=0, color=[255, 128, 0]),
+        1: dict(
+            link=('right_elbow', 'right_shoulder'), id=1, color=[255, 128, 0]),
+        2: dict(link=('right_shoulder', 'neck'), id=2, color=[51, 153, 255]),
+        3: dict(link=('neck', 'left_shoulder'), id=3, color=[51, 153, 255]),
+        4: dict(link=('left_shoulder', 'left_elbow'), id=4, color=[0, 255, 0]),
+        5: dict(link=('left_elbow', 'left_wrist'), id=5, color=[0, 255, 0]),
+        6: dict(link=('right_ankle', 'right_knee'), id=6, color=[255, 128, 0]),
+        7: dict(link=('right_knee', 'right_hip'), id=7, color=[255, 128, 0]),
+        8: dict(link=('right_hip', 'left_hip'), id=8, color=[51, 153, 255]),
+        9: dict(link=('left_hip', 'left_knee'), id=9, color=[0, 255, 0]),
+        10: dict(link=('left_knee', 'left_ankle'), id=10, color=[0, 255, 0]),
+        11: dict(link=('head_top', 'neck'), id=11, color=[51, 153, 255]),
+        12: dict(
+            link=('right_shoulder', 'right_hip'), id=12, color=[51, 153, 255]),
+        13:
+        dict(link=('left_shoulder', 'left_hip'), id=13, color=[51, 153, 255])
+    },
+    joint_weights=[
+        1., 1.2, 1.5, 1., 1.2, 1.5, 1., 1.2, 1.5, 1., 1.2, 1.5, 1., 1.
+    ],
+
+    # 'https://github.com/AIChallenger/AI_Challenger_2017/blob/master/'
+    # 'Evaluation/keypoint_eval/keypoint_eval.py#L50'
+    # delta = 2 x sigma
+    sigmas=[
+        0.01388152, 0.01515228, 0.01057665, 0.01417709, 0.01497891, 0.01402144,
+        0.03909642, 0.03686941, 0.01981803, 0.03843971, 0.03412318, 0.02415081,
+        0.01291456, 0.01236173
+    ])

configs/_base_/datasets/animalpose.py

@@ -0,0 +1,166 @@
+dataset_info = dict(
+    dataset_name='animalpose',
+    paper_info=dict(
+        author='Cao, Jinkun and Tang, Hongyang and Fang, Hao-Shu and '
+        'Shen, Xiaoyong and Lu, Cewu and Tai, Yu-Wing',
+        title='Cross-Domain Adaptation for Animal Pose Estimation',
+        container='The IEEE International Conference on '
+        'Computer Vision (ICCV)',
+        year='2019',
+        homepage='https://sites.google.com/view/animal-pose/',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='L_Eye', id=0, color=[0, 255, 0], type='upper', swap='R_Eye'),
+        1:
+        dict(
+            name='R_Eye',
+            id=1,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_Eye'),
+        2:
+        dict(
+            name='L_EarBase',
+            id=2,
+            color=[0, 255, 0],
+            type='upper',
+            swap='R_EarBase'),
+        3:
+        dict(
+            name='R_EarBase',
+            id=3,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_EarBase'),
+        4:
+        dict(name='Nose', id=4, color=[51, 153, 255], type='upper', swap=''),
+        5:
+        dict(name='Throat', id=5, color=[51, 153, 255], type='upper', swap=''),
+        6:
+        dict(
+            name='TailBase', id=6, color=[51, 153, 255], type='lower',
+            swap=''),
+        7:
+        dict(
+            name='Withers', id=7, color=[51, 153, 255], type='upper', swap=''),
+        8:
+        dict(
+            name='L_F_Elbow',
+            id=8,
+            color=[0, 255, 0],
+            type='upper',
+            swap='R_F_Elbow'),
+        9:
+        dict(
+            name='R_F_Elbow',
+            id=9,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_F_Elbow'),
+        10:
+        dict(
+            name='L_B_Elbow',
+            id=10,
+            color=[0, 255, 0],
+            type='lower',
+            swap='R_B_Elbow'),
+        11:
+        dict(
+            name='R_B_Elbow',
+            id=11,
+            color=[255, 128, 0],
+            type='lower',
+            swap='L_B_Elbow'),
+        12:
+        dict(
+            name='L_F_Knee',
+            id=12,
+            color=[0, 255, 0],
+            type='upper',
+            swap='R_F_Knee'),
+        13:
+        dict(
+            name='R_F_Knee',
+            id=13,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_F_Knee'),
+        14:
+        dict(
+            name='L_B_Knee',
+            id=14,
+            color=[0, 255, 0],
+            type='lower',
+            swap='R_B_Knee'),
+        15:
+        dict(
+            name='R_B_Knee',
+            id=15,
+            color=[255, 128, 0],
+            type='lower',
+            swap='L_B_Knee'),
+        16:
+        dict(
+            name='L_F_Paw',
+            id=16,
+            color=[0, 255, 0],
+            type='upper',
+            swap='R_F_Paw'),
+        17:
+        dict(
+            name='R_F_Paw',
+            id=17,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_F_Paw'),
+        18:
+        dict(
+            name='L_B_Paw',
+            id=18,
+            color=[0, 255, 0],
+            type='lower',
+            swap='R_B_Paw'),
+        19:
+        dict(
+            name='R_B_Paw',
+            id=19,
+            color=[255, 128, 0],
+            type='lower',
+            swap='L_B_Paw')
+    },
+    skeleton_info={
+        0: dict(link=('L_Eye', 'R_Eye'), id=0, color=[51, 153, 255]),
+        1: dict(link=('L_Eye', 'L_EarBase'), id=1, color=[0, 255, 0]),
+        2: dict(link=('R_Eye', 'R_EarBase'), id=2, color=[255, 128, 0]),
+        3: dict(link=('L_Eye', 'Nose'), id=3, color=[0, 255, 0]),
+        4: dict(link=('R_Eye', 'Nose'), id=4, color=[255, 128, 0]),
+        5: dict(link=('Nose', 'Throat'), id=5, color=[51, 153, 255]),
+        6: dict(link=('Throat', 'Withers'), id=6, color=[51, 153, 255]),
+        7: dict(link=('TailBase', 'Withers'), id=7, color=[51, 153, 255]),
+        8: dict(link=('Throat', 'L_F_Elbow'), id=8, color=[0, 255, 0]),
+        9: dict(link=('L_F_Elbow', 'L_F_Knee'), id=9, color=[0, 255, 0]),
+        10: dict(link=('L_F_Knee', 'L_F_Paw'), id=10, color=[0, 255, 0]),
+        11: dict(link=('Throat', 'R_F_Elbow'), id=11, color=[255, 128, 0]),
+        12: dict(link=('R_F_Elbow', 'R_F_Knee'), id=12, color=[255, 128, 0]),
+        13: dict(link=('R_F_Knee', 'R_F_Paw'), id=13, color=[255, 128, 0]),
+        14: dict(link=('TailBase', 'L_B_Elbow'), id=14, color=[0, 255, 0]),
+        15: dict(link=('L_B_Elbow', 'L_B_Knee'), id=15, color=[0, 255, 0]),
+        16: dict(link=('L_B_Knee', 'L_B_Paw'), id=16, color=[0, 255, 0]),
+        17: dict(link=('TailBase', 'R_B_Elbow'), id=17, color=[255, 128, 0]),
+        18: dict(link=('R_B_Elbow', 'R_B_Knee'), id=18, color=[255, 128, 0]),
+        19: dict(link=('R_B_Knee', 'R_B_Paw'), id=19, color=[255, 128, 0])
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.2, 1.2,
+        1.5, 1.5, 1.5, 1.5
+    ],
+
+    # Note: The original paper did not provide enough information about
+    # the sigmas. We modified from 'https://github.com/cocodataset/'
+    # 'cocoapi/blob/master/PythonAPI/pycocotools/cocoeval.py#L523'
+    sigmas=[
+        0.025, 0.025, 0.026, 0.035, 0.035, 0.10, 0.10, 0.10, 0.107, 0.107,
+        0.107, 0.107, 0.087, 0.087, 0.087, 0.087, 0.089, 0.089, 0.089, 0.089
+    ])

configs/_base_/datasets/ap10k.py

@@ -0,0 +1,142 @@
+dataset_info = dict(
+    dataset_name='ap10k',
+    paper_info=dict(
+        author='Yu, Hang and Xu, Yufei and Zhang, Jing and '
+        'Zhao, Wei and Guan, Ziyu and Tao, Dacheng',
+        title='AP-10K: A Benchmark for Animal Pose Estimation in the Wild',
+        container='35th Conference on Neural Information Processing Systems '
+        '(NeurIPS 2021) Track on Datasets and Bench-marks.',
+        year='2021',
+        homepage='https://github.com/AlexTheBad/AP-10K',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='L_Eye', id=0, color=[0, 255, 0], type='upper', swap='R_Eye'),
+        1:
+        dict(
+            name='R_Eye',
+            id=1,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_Eye'),
+        2:
+        dict(name='Nose', id=2, color=[51, 153, 255], type='upper', swap=''),
+        3:
+        dict(name='Neck', id=3, color=[51, 153, 255], type='upper', swap=''),
+        4:
+        dict(
+            name='Root of tail',
+            id=4,
+            color=[51, 153, 255],
+            type='lower',
+            swap=''),
+        5:
+        dict(
+            name='L_Shoulder',
+            id=5,
+            color=[51, 153, 255],
+            type='upper',
+            swap='R_Shoulder'),
+        6:
+        dict(
+            name='L_Elbow',
+            id=6,
+            color=[51, 153, 255],
+            type='upper',
+            swap='R_Elbow'),
+        7:
+        dict(
+            name='L_F_Paw',
+            id=7,
+            color=[0, 255, 0],
+            type='upper',
+            swap='R_F_Paw'),
+        8:
+        dict(
+            name='R_Shoulder',
+            id=8,
+            color=[0, 255, 0],
+            type='upper',
+            swap='L_Shoulder'),
+        9:
+        dict(
+            name='R_Elbow',
+            id=9,
+            color=[255, 128, 0],
+            type='upper',
+            swap='L_Elbow'),
+        10:
+        dict(
+            name='R_F_Paw',
+            id=10,
+            color=[0, 255, 0],
+            type='lower',
+            swap='L_F_Paw'),
+        11:
+        dict(
+            name='L_Hip',
+            id=11,
+            color=[255, 128, 0],
+            type='lower',
+            swap='R_Hip'),
+        12:
+        dict(
+            name='L_Knee',
+            id=12,
+            color=[255, 128, 0],
+            type='lower',
+            swap='R_Knee'),
+        13:
+        dict(
+            name='L_B_Paw',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='R_B_Paw'),
+        14:
+        dict(
+            name='R_Hip', id=14, color=[0, 255, 0], type='lower',
+            swap='L_Hip'),
+        15:
+        dict(
+            name='R_Knee',
+            id=15,
+            color=[0, 255, 0],
+            type='lower',
+            swap='L_Knee'),
+        16:
+        dict(
+            name='R_B_Paw',
+            id=16,
+            color=[0, 255, 0],
+            type='lower',
+            swap='L_B_Paw'),
+    },
+    skeleton_info={
+        0: dict(link=('L_Eye', 'R_Eye'), id=0, color=[0, 0, 255]),
+        1: dict(link=('L_Eye', 'Nose'), id=1, color=[0, 0, 255]),
+        2: dict(link=('R_Eye', 'Nose'), id=2, color=[0, 0, 255]),
+        3: dict(link=('Nose', 'Neck'), id=3, color=[0, 255, 0]),
+        4: dict(link=('Neck', 'Root of tail'), id=4, color=[0, 255, 0]),
+        5: dict(link=('Neck', 'L_Shoulder'), id=5, color=[0, 255, 255]),
+        6: dict(link=('L_Shoulder', 'L_Elbow'), id=6, color=[0, 255, 255]),
+        7: dict(link=('L_Elbow', 'L_F_Paw'), id=6, color=[0, 255, 255]),
+        8: dict(link=('Neck', 'R_Shoulder'), id=7, color=[6, 156, 250]),
+        9: dict(link=('R_Shoulder', 'R_Elbow'), id=8, color=[6, 156, 250]),
+        10: dict(link=('R_Elbow', 'R_F_Paw'), id=9, color=[6, 156, 250]),
+        11: dict(link=('Root of tail', 'L_Hip'), id=10, color=[0, 255, 255]),
+        12: dict(link=('L_Hip', 'L_Knee'), id=11, color=[0, 255, 255]),
+        13: dict(link=('L_Knee', 'L_B_Paw'), id=12, color=[0, 255, 255]),
+        14: dict(link=('Root of tail', 'R_Hip'), id=13, color=[6, 156, 250]),
+        15: dict(link=('R_Hip', 'R_Knee'), id=14, color=[6, 156, 250]),
+        16: dict(link=('R_Knee', 'R_B_Paw'), id=15, color=[6, 156, 250]),
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.5, 1.5, 1., 1., 1.2, 1.2, 1.5,
+        1.5
+    ],
+    sigmas=[
+        0.025, 0.025, 0.026, 0.035, 0.035, 0.079, 0.072, 0.062, 0.079, 0.072,
+        0.062, 0.107, 0.087, 0.089, 0.107, 0.087, 0.089
+    ])

configs/_base_/datasets/atrw.py

@@ -0,0 +1,144 @@
+dataset_info = dict(
+    dataset_name='atrw',
+    paper_info=dict(
+        author='Li, Shuyuan and Li, Jianguo and Tang, Hanlin '
+        'and Qian, Rui and Lin, Weiyao',
+        title='ATRW: A Benchmark for Amur Tiger '
+        'Re-identification in the Wild',
+        container='Proceedings of the 28th ACM '
+        'International Conference on Multimedia',
+        year='2020',
+        homepage='https://cvwc2019.github.io/challenge.html',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='left_ear',
+            id=0,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_ear'),
+        1:
+        dict(
+            name='right_ear',
+            id=1,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_ear'),
+        2:
+        dict(name='nose', id=2, color=[51, 153, 255], type='upper', swap=''),
+        3:
+        dict(
+            name='right_shoulder',
+            id=3,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        4:
+        dict(
+            name='right_front_paw',
+            id=4,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_front_paw'),
+        5:
+        dict(
+            name='left_shoulder',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        6:
+        dict(
+            name='left_front_paw',
+            id=6,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_front_paw'),
+        7:
+        dict(
+            name='right_hip',
+            id=7,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        8:
+        dict(
+            name='right_knee',
+            id=8,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        9:
+        dict(
+            name='right_back_paw',
+            id=9,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_back_paw'),
+        10:
+        dict(
+            name='left_hip',
+            id=10,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        11:
+        dict(
+            name='left_knee',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        12:
+        dict(
+            name='left_back_paw',
+            id=12,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_back_paw'),
+        13:
+        dict(name='tail', id=13, color=[51, 153, 255], type='lower', swap=''),
+        14:
+        dict(
+            name='center', id=14, color=[51, 153, 255], type='lower', swap=''),
+    },
+    skeleton_info={
+        0:
+        dict(link=('left_ear', 'nose'), id=0, color=[51, 153, 255]),
+        1:
+        dict(link=('right_ear', 'nose'), id=1, color=[51, 153, 255]),
+        2:
+        dict(link=('nose', 'center'), id=2, color=[51, 153, 255]),
+        3:
+        dict(
+            link=('left_shoulder', 'left_front_paw'), id=3, color=[0, 255, 0]),
+        4:
+        dict(link=('left_shoulder', 'center'), id=4, color=[0, 255, 0]),
+        5:
+        dict(
+            link=('right_shoulder', 'right_front_paw'),
+            id=5,
+            color=[255, 128, 0]),
+        6:
+        dict(link=('right_shoulder', 'center'), id=6, color=[255, 128, 0]),
+        7:
+        dict(link=('tail', 'center'), id=7, color=[51, 153, 255]),
+        8:
+        dict(link=('right_back_paw', 'right_knee'), id=8, color=[255, 128, 0]),
+        9:
+        dict(link=('right_knee', 'right_hip'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('right_hip', 'tail'), id=10, color=[255, 128, 0]),
+        11:
+        dict(link=('left_back_paw', 'left_knee'), id=11, color=[0, 255, 0]),
+        12:
+        dict(link=('left_knee', 'left_hip'), id=12, color=[0, 255, 0]),
+        13:
+        dict(link=('left_hip', 'tail'), id=13, color=[0, 255, 0]),
+    },
+    joint_weights=[1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.],
+    sigmas=[
+        0.0277, 0.0823, 0.0831, 0.0202, 0.0716, 0.0263, 0.0646, 0.0302, 0.0440,
+        0.0316, 0.0333, 0.0547, 0.0263, 0.0683, 0.0539
+    ])

configs/_base_/datasets/coco.py

@@ -0,0 +1,181 @@
+dataset_info = dict(
+    dataset_name='coco',
+    paper_info=dict(
+        author='Lin, Tsung-Yi and Maire, Michael and '
+        'Belongie, Serge and Hays, James and '
+        'Perona, Pietro and Ramanan, Deva and '
+        r'Doll{\'a}r, Piotr and Zitnick, C Lawrence',
+        title='Microsoft coco: Common objects in context',
+        container='European conference on computer vision',
+        year='2014',
+        homepage='http://cocodataset.org/',
+    ),
+    keypoint_info={
+        0:
+        dict(name='nose', id=0, color=[51, 153, 255], type='upper', swap=''),
+        1:
+        dict(
+            name='left_eye',
+            id=1,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_eye'),
+        2:
+        dict(
+            name='right_eye',
+            id=2,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_eye'),
+        3:
+        dict(
+            name='left_ear',
+            id=3,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_ear'),
+        4:
+        dict(
+            name='right_ear',
+            id=4,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_ear'),
+        5:
+        dict(
+            name='left_shoulder',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        6:
+        dict(
+            name='right_shoulder',
+            id=6,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        7:
+        dict(
+            name='left_elbow',
+            id=7,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        8:
+        dict(
+            name='right_elbow',
+            id=8,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        9:
+        dict(
+            name='left_wrist',
+            id=9,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        10:
+        dict(
+            name='right_wrist',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        11:
+        dict(
+            name='left_hip',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        12:
+        dict(
+            name='right_hip',
+            id=12,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        13:
+        dict(
+            name='left_knee',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        14:
+        dict(
+            name='right_knee',
+            id=14,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        15:
+        dict(
+            name='left_ankle',
+            id=15,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        16:
+        dict(
+            name='right_ankle',
+            id=16,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle')
+    },
+    skeleton_info={
+        0:
+        dict(link=('left_ankle', 'left_knee'), id=0, color=[0, 255, 0]),
+        1:
+        dict(link=('left_knee', 'left_hip'), id=1, color=[0, 255, 0]),
+        2:
+        dict(link=('right_ankle', 'right_knee'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('right_knee', 'right_hip'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('left_hip', 'right_hip'), id=4, color=[51, 153, 255]),
+        5:
+        dict(link=('left_shoulder', 'left_hip'), id=5, color=[51, 153, 255]),
+        6:
+        dict(link=('right_shoulder', 'right_hip'), id=6, color=[51, 153, 255]),
+        7:
+        dict(
+            link=('left_shoulder', 'right_shoulder'),
+            id=7,
+            color=[51, 153, 255]),
+        8:
+        dict(link=('left_shoulder', 'left_elbow'), id=8, color=[0, 255, 0]),
+        9:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('left_elbow', 'left_wrist'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('left_eye', 'right_eye'), id=12, color=[51, 153, 255]),
+        13:
+        dict(link=('nose', 'left_eye'), id=13, color=[51, 153, 255]),
+        14:
+        dict(link=('nose', 'right_eye'), id=14, color=[51, 153, 255]),
+        15:
+        dict(link=('left_eye', 'left_ear'), id=15, color=[51, 153, 255]),
+        16:
+        dict(link=('right_eye', 'right_ear'), id=16, color=[51, 153, 255]),
+        17:
+        dict(link=('left_ear', 'left_shoulder'), id=17, color=[51, 153, 255]),
+        18:
+        dict(
+            link=('right_ear', 'right_shoulder'), id=18, color=[51, 153, 255])
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.5, 1.5, 1., 1., 1.2, 1.2, 1.5,
+        1.5
+    ],
+    sigmas=[
+        0.026, 0.025, 0.025, 0.035, 0.035, 0.079, 0.079, 0.072, 0.072, 0.062,
+        0.062, 0.107, 0.107, 0.087, 0.087, 0.089, 0.089
+    ])

configs/_base_/datasets/coco_wholebody_face.py

@@ -0,0 +1,448 @@
+dataset_info = dict(
+    dataset_name='coco_wholebody_face',
+    paper_info=dict(
+        author='Jin, Sheng and Xu, Lumin and Xu, Jin and '
+        'Wang, Can and Liu, Wentao and '
+        'Qian, Chen and Ouyang, Wanli and Luo, Ping',
+        title='Whole-Body Human Pose Estimation in the Wild',
+        container='Proceedings of the European '
+        'Conference on Computer Vision (ECCV)',
+        year='2020',
+        homepage='https://github.com/jin-s13/COCO-WholeBody/',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='face-0',
+            id=0,
+            color=[255, 255, 255],
+            type='',
+            swap='face-16'),
+        1:
+        dict(
+            name='face-1',
+            id=1,
+            color=[255, 255, 255],
+            type='',
+            swap='face-15'),
+        2:
+        dict(
+            name='face-2',
+            id=2,
+            color=[255, 255, 255],
+            type='',
+            swap='face-14'),
+        3:
+        dict(
+            name='face-3',
+            id=3,
+            color=[255, 255, 255],
+            type='',
+            swap='face-13'),
+        4:
+        dict(
+            name='face-4',
+            id=4,
+            color=[255, 255, 255],
+            type='',
+            swap='face-12'),
+        5:
+        dict(
+            name='face-5',
+            id=5,
+            color=[255, 255, 255],
+            type='',
+            swap='face-11'),
+        6:
+        dict(
+            name='face-6',
+            id=6,
+            color=[255, 255, 255],
+            type='',
+            swap='face-10'),
+        7:
+        dict(
+            name='face-7', id=7, color=[255, 255, 255], type='',
+            swap='face-9'),
+        8:
+        dict(name='face-8', id=8, color=[255, 255, 255], type='', swap=''),
+        9:
+        dict(
+            name='face-9', id=9, color=[255, 255, 255], type='',
+            swap='face-7'),
+        10:
+        dict(
+            name='face-10',
+            id=10,
+            color=[255, 255, 255],
+            type='',
+            swap='face-6'),
+        11:
+        dict(
+            name='face-11',
+            id=11,
+            color=[255, 255, 255],
+            type='',
+            swap='face-5'),
+        12:
+        dict(
+            name='face-12',
+            id=12,
+            color=[255, 255, 255],
+            type='',
+            swap='face-4'),
+        13:
+        dict(
+            name='face-13',
+            id=13,
+            color=[255, 255, 255],
+            type='',
+            swap='face-3'),
+        14:
+        dict(
+            name='face-14',
+            id=14,
+            color=[255, 255, 255],
+            type='',
+            swap='face-2'),
+        15:
+        dict(
+            name='face-15',
+            id=15,
+            color=[255, 255, 255],
+            type='',
+            swap='face-1'),
+        16:
+        dict(
+            name='face-16',
+            id=16,
+            color=[255, 255, 255],
+            type='',
+            swap='face-0'),
+        17:
+        dict(
+            name='face-17',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='face-26'),
+        18:
+        dict(
+            name='face-18',
+            id=18,
+            color=[255, 255, 255],
+            type='',
+            swap='face-25'),
+        19:
+        dict(
+            name='face-19',
+            id=19,
+            color=[255, 255, 255],
+            type='',
+            swap='face-24'),
+        20:
+        dict(
+            name='face-20',
+            id=20,
+            color=[255, 255, 255],
+            type='',
+            swap='face-23'),
+        21:
+        dict(
+            name='face-21',
+            id=21,
+            color=[255, 255, 255],
+            type='',
+            swap='face-22'),
+        22:
+        dict(
+            name='face-22',
+            id=22,
+            color=[255, 255, 255],
+            type='',
+            swap='face-21'),
+        23:
+        dict(
+            name='face-23',
+            id=23,
+            color=[255, 255, 255],
+            type='',
+            swap='face-20'),
+        24:
+        dict(
+            name='face-24',
+            id=24,
+            color=[255, 255, 255],
+            type='',
+            swap='face-19'),
+        25:
+        dict(
+            name='face-25',
+            id=25,
+            color=[255, 255, 255],
+            type='',
+            swap='face-18'),
+        26:
+        dict(
+            name='face-26',
+            id=26,
+            color=[255, 255, 255],
+            type='',
+            swap='face-17'),
+        27:
+        dict(name='face-27', id=27, color=[255, 255, 255], type='', swap=''),
+        28:
+        dict(name='face-28', id=28, color=[255, 255, 255], type='', swap=''),
+        29:
+        dict(name='face-29', id=29, color=[255, 255, 255], type='', swap=''),
+        30:
+        dict(name='face-30', id=30, color=[255, 255, 255], type='', swap=''),
+        31:
+        dict(
+            name='face-31',
+            id=31,
+            color=[255, 255, 255],
+            type='',
+            swap='face-35'),
+        32:
+        dict(
+            name='face-32',
+            id=32,
+            color=[255, 255, 255],
+            type='',
+            swap='face-34'),
+        33:
+        dict(name='face-33', id=33, color=[255, 255, 255], type='', swap=''),
+        34:
+        dict(
+            name='face-34',
+            id=34,
+            color=[255, 255, 255],
+            type='',
+            swap='face-32'),
+        35:
+        dict(
+            name='face-35',
+            id=35,
+            color=[255, 255, 255],
+            type='',
+            swap='face-31'),
+        36:
+        dict(
+            name='face-36',
+            id=36,
+            color=[255, 255, 255],
+            type='',
+            swap='face-45'),
+        37:
+        dict(
+            name='face-37',
+            id=37,
+            color=[255, 255, 255],
+            type='',
+            swap='face-44'),
+        38:
+        dict(
+            name='face-38',
+            id=38,
+            color=[255, 255, 255],
+            type='',
+            swap='face-43'),
+        39:
+        dict(
+            name='face-39',
+            id=39,
+            color=[255, 255, 255],
+            type='',
+            swap='face-42'),
+        40:
+        dict(
+            name='face-40',
+            id=40,
+            color=[255, 255, 255],
+            type='',
+            swap='face-47'),
+        41:
+        dict(
+            name='face-41',
+            id=41,
+            color=[255, 255, 255],
+            type='',
+            swap='face-46'),
+        42:
+        dict(
+            name='face-42',
+            id=42,
+            color=[255, 255, 255],
+            type='',
+            swap='face-39'),
+        43:
+        dict(
+            name='face-43',
+            id=43,
+            color=[255, 255, 255],
+            type='',
+            swap='face-38'),
+        44:
+        dict(
+            name='face-44',
+            id=44,
+            color=[255, 255, 255],
+            type='',
+            swap='face-37'),
+        45:
+        dict(
+            name='face-45',
+            id=45,
+            color=[255, 255, 255],
+            type='',
+            swap='face-36'),
+        46:
+        dict(
+            name='face-46',
+            id=46,
+            color=[255, 255, 255],
+            type='',
+            swap='face-41'),
+        47:
+        dict(
+            name='face-47',
+            id=47,
+            color=[255, 255, 255],
+            type='',
+            swap='face-40'),
+        48:
+        dict(
+            name='face-48',
+            id=48,
+            color=[255, 255, 255],
+            type='',
+            swap='face-54'),
+        49:
+        dict(
+            name='face-49',
+            id=49,
+            color=[255, 255, 255],
+            type='',
+            swap='face-53'),
+        50:
+        dict(
+            name='face-50',
+            id=50,
+            color=[255, 255, 255],
+            type='',
+            swap='face-52'),
+        51:
+        dict(name='face-51', id=52, color=[255, 255, 255], type='', swap=''),
+        52:
+        dict(
+            name='face-52',
+            id=52,
+            color=[255, 255, 255],
+            type='',
+            swap='face-50'),
+        53:
+        dict(
+            name='face-53',
+            id=53,
+            color=[255, 255, 255],
+            type='',
+            swap='face-49'),
+        54:
+        dict(
+            name='face-54',
+            id=54,
+            color=[255, 255, 255],
+            type='',
+            swap='face-48'),
+        55:
+        dict(
+            name='face-55',
+            id=55,
+            color=[255, 255, 255],
+            type='',
+            swap='face-59'),
+        56:
+        dict(
+            name='face-56',
+            id=56,
+            color=[255, 255, 255],
+            type='',
+            swap='face-58'),
+        57:
+        dict(name='face-57', id=57, color=[255, 255, 255], type='', swap=''),
+        58:
+        dict(
+            name='face-58',
+            id=58,
+            color=[255, 255, 255],
+            type='',
+            swap='face-56'),
+        59:
+        dict(
+            name='face-59',
+            id=59,
+            color=[255, 255, 255],
+            type='',
+            swap='face-55'),
+        60:
+        dict(
+            name='face-60',
+            id=60,
+            color=[255, 255, 255],
+            type='',
+            swap='face-64'),
+        61:
+        dict(
+            name='face-61',
+            id=61,
+            color=[255, 255, 255],
+            type='',
+            swap='face-63'),
+        62:
+        dict(name='face-62', id=62, color=[255, 255, 255], type='', swap=''),
+        63:
+        dict(
+            name='face-63',
+            id=63,
+            color=[255, 255, 255],
+            type='',
+            swap='face-61'),
+        64:
+        dict(
+            name='face-64',
+            id=64,
+            color=[255, 255, 255],
+            type='',
+            swap='face-60'),
+        65:
+        dict(
+            name='face-65',
+            id=65,
+            color=[255, 255, 255],
+            type='',
+            swap='face-67'),
+        66:
+        dict(name='face-66', id=66, color=[255, 255, 255], type='', swap=''),
+        67:
+        dict(
+            name='face-67',
+            id=67,
+            color=[255, 255, 255],
+            type='',
+            swap='face-65')
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 68,
+
+    # 'https://github.com/jin-s13/COCO-WholeBody/blob/master/'
+    # 'evaluation/myeval_wholebody.py#L177'
+    sigmas=[
+        0.042, 0.043, 0.044, 0.043, 0.040, 0.035, 0.031, 0.025, 0.020, 0.023,
+        0.029, 0.032, 0.037, 0.038, 0.043, 0.041, 0.045, 0.013, 0.012, 0.011,
+        0.011, 0.012, 0.012, 0.011, 0.011, 0.013, 0.015, 0.009, 0.007, 0.007,
+        0.007, 0.012, 0.009, 0.008, 0.016, 0.010, 0.017, 0.011, 0.009, 0.011,
+        0.009, 0.007, 0.013, 0.008, 0.011, 0.012, 0.010, 0.034, 0.008, 0.008,
+        0.009, 0.008, 0.008, 0.007, 0.010, 0.008, 0.009, 0.009, 0.009, 0.007,
+        0.007, 0.008, 0.011, 0.008, 0.008, 0.008, 0.01, 0.008
+    ])

configs/_base_/datasets/coco_wholebody_hand.py

@@ -0,0 +1,147 @@
+dataset_info = dict(
+    dataset_name='coco_wholebody_hand',
+    paper_info=dict(
+        author='Jin, Sheng and Xu, Lumin and Xu, Jin and '
+        'Wang, Can and Liu, Wentao and '
+        'Qian, Chen and Ouyang, Wanli and Luo, Ping',
+        title='Whole-Body Human Pose Estimation in the Wild',
+        container='Proceedings of the European '
+        'Conference on Computer Vision (ECCV)',
+        year='2020',
+        homepage='https://github.com/jin-s13/COCO-WholeBody/',
+    ),
+    keypoint_info={
+        0:
+        dict(name='wrist', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='thumb1', id=1, color=[255, 128, 0], type='', swap=''),
+        2:
+        dict(name='thumb2', id=2, color=[255, 128, 0], type='', swap=''),
+        3:
+        dict(name='thumb3', id=3, color=[255, 128, 0], type='', swap=''),
+        4:
+        dict(name='thumb4', id=4, color=[255, 128, 0], type='', swap=''),
+        5:
+        dict(
+            name='forefinger1', id=5, color=[255, 153, 255], type='', swap=''),
+        6:
+        dict(
+            name='forefinger2', id=6, color=[255, 153, 255], type='', swap=''),
+        7:
+        dict(
+            name='forefinger3', id=7, color=[255, 153, 255], type='', swap=''),
+        8:
+        dict(
+            name='forefinger4', id=8, color=[255, 153, 255], type='', swap=''),
+        9:
+        dict(
+            name='middle_finger1',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        10:
+        dict(
+            name='middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        11:
+        dict(
+            name='middle_finger3',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        12:
+        dict(
+            name='middle_finger4',
+            id=12,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        13:
+        dict(
+            name='ring_finger1', id=13, color=[255, 51, 51], type='', swap=''),
+        14:
+        dict(
+            name='ring_finger2', id=14, color=[255, 51, 51], type='', swap=''),
+        15:
+        dict(
+            name='ring_finger3', id=15, color=[255, 51, 51], type='', swap=''),
+        16:
+        dict(
+            name='ring_finger4', id=16, color=[255, 51, 51], type='', swap=''),
+        17:
+        dict(name='pinky_finger1', id=17, color=[0, 255, 0], type='', swap=''),
+        18:
+        dict(name='pinky_finger2', id=18, color=[0, 255, 0], type='', swap=''),
+        19:
+        dict(name='pinky_finger3', id=19, color=[0, 255, 0], type='', swap=''),
+        20:
+        dict(name='pinky_finger4', id=20, color=[0, 255, 0], type='', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('wrist', 'thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('thumb1', 'thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('thumb2', 'thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('thumb3', 'thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('wrist', 'forefinger1'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('forefinger1', 'forefinger2'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('forefinger2', 'forefinger3'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('forefinger3', 'forefinger4'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('wrist', 'middle_finger1'), id=8, color=[102, 178, 255]),
+        9:
+        dict(
+            link=('middle_finger1', 'middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('middle_finger2', 'middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('middle_finger3', 'middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(link=('wrist', 'ring_finger1'), id=12, color=[255, 51, 51]),
+        13:
+        dict(
+            link=('ring_finger1', 'ring_finger2'), id=13, color=[255, 51, 51]),
+        14:
+        dict(
+            link=('ring_finger2', 'ring_finger3'), id=14, color=[255, 51, 51]),
+        15:
+        dict(
+            link=('ring_finger3', 'ring_finger4'), id=15, color=[255, 51, 51]),
+        16:
+        dict(link=('wrist', 'pinky_finger1'), id=16, color=[0, 255, 0]),
+        17:
+        dict(
+            link=('pinky_finger1', 'pinky_finger2'), id=17, color=[0, 255, 0]),
+        18:
+        dict(
+            link=('pinky_finger2', 'pinky_finger3'), id=18, color=[0, 255, 0]),
+        19:
+        dict(
+            link=('pinky_finger3', 'pinky_finger4'), id=19, color=[0, 255, 0])
+    },
+    joint_weights=[1.] * 21,
+    sigmas=[
+        0.029, 0.022, 0.035, 0.037, 0.047, 0.026, 0.025, 0.024, 0.035, 0.018,
+        0.024, 0.022, 0.026, 0.017, 0.021, 0.021, 0.032, 0.02, 0.019, 0.022,
+        0.031
+    ])

configs/_base_/datasets/cofw.py

@@ -0,0 +1,134 @@
+dataset_info = dict(
+    dataset_name='cofw',
+    paper_info=dict(
+        author='Burgos-Artizzu, Xavier P and Perona, '
+        r'Pietro and Doll{\'a}r, Piotr',
+        title='Robust face landmark estimation under occlusion',
+        container='Proceedings of the IEEE international '
+        'conference on computer vision',
+        year='2013',
+        homepage='http://www.vision.caltech.edu/xpburgos/ICCV13/',
+    ),
+    keypoint_info={
+        0:
+        dict(name='kpt-0', id=0, color=[255, 255, 255], type='', swap='kpt-1'),
+        1:
+        dict(name='kpt-1', id=1, color=[255, 255, 255], type='', swap='kpt-0'),
+        2:
+        dict(name='kpt-2', id=2, color=[255, 255, 255], type='', swap='kpt-3'),
+        3:
+        dict(name='kpt-3', id=3, color=[255, 255, 255], type='', swap='kpt-2'),
+        4:
+        dict(name='kpt-4', id=4, color=[255, 255, 255], type='', swap='kpt-6'),
+        5:
+        dict(name='kpt-5', id=5, color=[255, 255, 255], type='', swap='kpt-7'),
+        6:
+        dict(name='kpt-6', id=6, color=[255, 255, 255], type='', swap='kpt-4'),
+        7:
+        dict(name='kpt-7', id=7, color=[255, 255, 255], type='', swap='kpt-5'),
+        8:
+        dict(name='kpt-8', id=8, color=[255, 255, 255], type='', swap='kpt-9'),
+        9:
+        dict(name='kpt-9', id=9, color=[255, 255, 255], type='', swap='kpt-8'),
+        10:
+        dict(
+            name='kpt-10',
+            id=10,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-11'),
+        11:
+        dict(
+            name='kpt-11',
+            id=11,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-10'),
+        12:
+        dict(
+            name='kpt-12',
+            id=12,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-14'),
+        13:
+        dict(
+            name='kpt-13',
+            id=13,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-15'),
+        14:
+        dict(
+            name='kpt-14',
+            id=14,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-12'),
+        15:
+        dict(
+            name='kpt-15',
+            id=15,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-13'),
+        16:
+        dict(
+            name='kpt-16',
+            id=16,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-17'),
+        17:
+        dict(
+            name='kpt-17',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-16'),
+        18:
+        dict(
+            name='kpt-18',
+            id=18,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-19'),
+        19:
+        dict(
+            name='kpt-19',
+            id=19,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-18'),
+        20:
+        dict(name='kpt-20', id=20, color=[255, 255, 255], type='', swap=''),
+        21:
+        dict(name='kpt-21', id=21, color=[255, 255, 255], type='', swap=''),
+        22:
+        dict(
+            name='kpt-22',
+            id=22,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-23'),
+        23:
+        dict(
+            name='kpt-23',
+            id=23,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-22'),
+        24:
+        dict(name='kpt-24', id=24, color=[255, 255, 255], type='', swap=''),
+        25:
+        dict(name='kpt-25', id=25, color=[255, 255, 255], type='', swap=''),
+        26:
+        dict(name='kpt-26', id=26, color=[255, 255, 255], type='', swap=''),
+        27:
+        dict(name='kpt-27', id=27, color=[255, 255, 255], type='', swap=''),
+        28:
+        dict(name='kpt-28', id=28, color=[255, 255, 255], type='', swap='')
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 29,
+    sigmas=[])

configs/_base_/datasets/crowdpose.py

@@ -0,0 +1,147 @@
+dataset_info = dict(
+    dataset_name='crowdpose',
+    paper_info=dict(
+        author='Li, Jiefeng and Wang, Can and Zhu, Hao and '
+        'Mao, Yihuan and Fang, Hao-Shu and Lu, Cewu',
+        title='CrowdPose: Efficient Crowded Scenes Pose Estimation '
+        'and A New Benchmark',
+        container='Proceedings of IEEE Conference on Computer '
+        'Vision and Pattern Recognition (CVPR)',
+        year='2019',
+        homepage='https://github.com/Jeff-sjtu/CrowdPose',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='left_shoulder',
+            id=0,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_shoulder'),
+        1:
+        dict(
+            name='right_shoulder',
+            id=1,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_shoulder'),
+        2:
+        dict(
+            name='left_elbow',
+            id=2,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_elbow'),
+        3:
+        dict(
+            name='right_elbow',
+            id=3,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_elbow'),
+        4:
+        dict(
+            name='left_wrist',
+            id=4,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_wrist'),
+        5:
+        dict(
+            name='right_wrist',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='left_wrist'),
+        6:
+        dict(
+            name='left_hip',
+            id=6,
+            color=[255, 128, 0],
+            type='lower',
+            swap='right_hip'),
+        7:
+        dict(
+            name='right_hip',
+            id=7,
+            color=[0, 255, 0],
+            type='lower',
+            swap='left_hip'),
+        8:
+        dict(
+            name='left_knee',
+            id=8,
+            color=[255, 128, 0],
+            type='lower',
+            swap='right_knee'),
+        9:
+        dict(
+            name='right_knee',
+            id=9,
+            color=[0, 255, 0],
+            type='lower',
+            swap='left_knee'),
+        10:
+        dict(
+            name='left_ankle',
+            id=10,
+            color=[255, 128, 0],
+            type='lower',
+            swap='right_ankle'),
+        11:
+        dict(
+            name='right_ankle',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='left_ankle'),
+        12:
+        dict(
+            name='top_head', id=12, color=[255, 128, 0], type='upper',
+            swap=''),
+        13:
+        dict(name='neck', id=13, color=[0, 255, 0], type='upper', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('left_ankle', 'left_knee'), id=0, color=[0, 255, 0]),
+        1:
+        dict(link=('left_knee', 'left_hip'), id=1, color=[0, 255, 0]),
+        2:
+        dict(link=('right_ankle', 'right_knee'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('right_knee', 'right_hip'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('left_hip', 'right_hip'), id=4, color=[51, 153, 255]),
+        5:
+        dict(link=('left_shoulder', 'left_hip'), id=5, color=[51, 153, 255]),
+        6:
+        dict(link=('right_shoulder', 'right_hip'), id=6, color=[51, 153, 255]),
+        7:
+        dict(
+            link=('left_shoulder', 'right_shoulder'),
+            id=7,
+            color=[51, 153, 255]),
+        8:
+        dict(link=('left_shoulder', 'left_elbow'), id=8, color=[0, 255, 0]),
+        9:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('left_elbow', 'left_wrist'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('top_head', 'neck'), id=12, color=[51, 153, 255]),
+        13:
+        dict(link=('right_shoulder', 'neck'), id=13, color=[51, 153, 255]),
+        14:
+        dict(link=('left_shoulder', 'neck'), id=14, color=[51, 153, 255])
+    },
+    joint_weights=[
+        0.2, 0.2, 0.2, 1.3, 1.5, 0.2, 1.3, 1.5, 0.2, 0.2, 0.5, 0.2, 0.2, 0.5
+    ],
+    sigmas=[
+        0.079, 0.079, 0.072, 0.072, 0.062, 0.062, 0.107, 0.107, 0.087, 0.087,
+        0.089, 0.089, 0.079, 0.079
+    ])

configs/_base_/datasets/deepfashion_full.py

@@ -0,0 +1,74 @@
+dataset_info = dict(
+    dataset_name='deepfashion_full',
+    paper_info=dict(
+        author='Liu, Ziwei and Luo, Ping and Qiu, Shi '
+        'and Wang, Xiaogang and Tang, Xiaoou',
+        title='DeepFashion: Powering Robust Clothes Recognition '
+        'and Retrieval with Rich Annotations',
+        container='Proceedings of IEEE Conference on Computer '
+        'Vision and Pattern Recognition (CVPR)',
+        year='2016',
+        homepage='http://mmlab.ie.cuhk.edu.hk/projects/'
+        'DeepFashion/LandmarkDetection.html',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='left collar',
+            id=0,
+            color=[255, 255, 255],
+            type='',
+            swap='right collar'),
+        1:
+        dict(
+            name='right collar',
+            id=1,
+            color=[255, 255, 255],
+            type='',
+            swap='left collar'),
+        2:
+        dict(
+            name='left sleeve',
+            id=2,
+            color=[255, 255, 255],
+            type='',
+            swap='right sleeve'),
+        3:
+        dict(
+            name='right sleeve',
+            id=3,
+            color=[255, 255, 255],
+            type='',
+            swap='left sleeve'),
+        4:
+        dict(
+            name='left waistline',
+            id=0,
+            color=[255, 255, 255],
+            type='',
+            swap='right waistline'),
+        5:
+        dict(
+            name='right waistline',
+            id=1,
+            color=[255, 255, 255],
+            type='',
+            swap='left waistline'),
+        6:
+        dict(
+            name='left hem',
+            id=2,
+            color=[255, 255, 255],
+            type='',
+            swap='right hem'),
+        7:
+        dict(
+            name='right hem',
+            id=3,
+            color=[255, 255, 255],
+            type='',
+            swap='left hem'),
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 8,
+    sigmas=[])

configs/_base_/datasets/deepfashion_lower.py

@@ -0,0 +1,46 @@
+dataset_info = dict(
+    dataset_name='deepfashion_lower',
+    paper_info=dict(
+        author='Liu, Ziwei and Luo, Ping and Qiu, Shi '
+        'and Wang, Xiaogang and Tang, Xiaoou',
+        title='DeepFashion: Powering Robust Clothes Recognition '
+        'and Retrieval with Rich Annotations',
+        container='Proceedings of IEEE Conference on Computer '
+        'Vision and Pattern Recognition (CVPR)',
+        year='2016',
+        homepage='http://mmlab.ie.cuhk.edu.hk/projects/'
+        'DeepFashion/LandmarkDetection.html',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='left waistline',
+            id=0,
+            color=[255, 255, 255],
+            type='',
+            swap='right waistline'),
+        1:
+        dict(
+            name='right waistline',
+            id=1,
+            color=[255, 255, 255],
+            type='',
+            swap='left waistline'),
+        2:
+        dict(
+            name='left hem',
+            id=2,
+            color=[255, 255, 255],
+            type='',
+            swap='right hem'),
+        3:
+        dict(
+            name='right hem',
+            id=3,
+            color=[255, 255, 255],
+            type='',
+            swap='left hem'),
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 4,
+    sigmas=[])

configs/_base_/datasets/deepfashion_upper.py

@@ -0,0 +1,60 @@
+dataset_info = dict(
+    dataset_name='deepfashion_upper',
+    paper_info=dict(
+        author='Liu, Ziwei and Luo, Ping and Qiu, Shi '
+        'and Wang, Xiaogang and Tang, Xiaoou',
+        title='DeepFashion: Powering Robust Clothes Recognition '
+        'and Retrieval with Rich Annotations',
+        container='Proceedings of IEEE Conference on Computer '
+        'Vision and Pattern Recognition (CVPR)',
+        year='2016',
+        homepage='http://mmlab.ie.cuhk.edu.hk/projects/'
+        'DeepFashion/LandmarkDetection.html',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='left collar',
+            id=0,
+            color=[255, 255, 255],
+            type='',
+            swap='right collar'),
+        1:
+        dict(
+            name='right collar',
+            id=1,
+            color=[255, 255, 255],
+            type='',
+            swap='left collar'),
+        2:
+        dict(
+            name='left sleeve',
+            id=2,
+            color=[255, 255, 255],
+            type='',
+            swap='right sleeve'),
+        3:
+        dict(
+            name='right sleeve',
+            id=3,
+            color=[255, 255, 255],
+            type='',
+            swap='left sleeve'),
+        4:
+        dict(
+            name='left hem',
+            id=4,
+            color=[255, 255, 255],
+            type='',
+            swap='right hem'),
+        5:
+        dict(
+            name='right hem',
+            id=5,
+            color=[255, 255, 255],
+            type='',
+            swap='left hem'),
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 6,
+    sigmas=[])

configs/_base_/datasets/fly.py

@@ -0,0 +1,237 @@
+dataset_info = dict(
+    dataset_name='fly',
+    paper_info=dict(
+        author='Pereira, Talmo D and Aldarondo, Diego E and '
+        'Willmore, Lindsay and Kislin, Mikhail and '
+        'Wang, Samuel S-H and Murthy, Mala and Shaevitz, Joshua W',
+        title='Fast animal pose estimation using deep neural networks',
+        container='Nature methods',
+        year='2019',
+        homepage='https://github.com/jgraving/DeepPoseKit-Data',
+    ),
+    keypoint_info={
+        0:
+        dict(name='head', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='eyeL', id=1, color=[255, 255, 255], type='', swap='eyeR'),
+        2:
+        dict(name='eyeR', id=2, color=[255, 255, 255], type='', swap='eyeL'),
+        3:
+        dict(name='neck', id=3, color=[255, 255, 255], type='', swap=''),
+        4:
+        dict(name='thorax', id=4, color=[255, 255, 255], type='', swap=''),
+        5:
+        dict(name='abdomen', id=5, color=[255, 255, 255], type='', swap=''),
+        6:
+        dict(
+            name='forelegR1',
+            id=6,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL1'),
+        7:
+        dict(
+            name='forelegR2',
+            id=7,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL2'),
+        8:
+        dict(
+            name='forelegR3',
+            id=8,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL3'),
+        9:
+        dict(
+            name='forelegR4',
+            id=9,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL4'),
+        10:
+        dict(
+            name='midlegR1',
+            id=10,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL1'),
+        11:
+        dict(
+            name='midlegR2',
+            id=11,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL2'),
+        12:
+        dict(
+            name='midlegR3',
+            id=12,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL3'),
+        13:
+        dict(
+            name='midlegR4',
+            id=13,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL4'),
+        14:
+        dict(
+            name='hindlegR1',
+            id=14,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL1'),
+        15:
+        dict(
+            name='hindlegR2',
+            id=15,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL2'),
+        16:
+        dict(
+            name='hindlegR3',
+            id=16,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL3'),
+        17:
+        dict(
+            name='hindlegR4',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL4'),
+        18:
+        dict(
+            name='forelegL1',
+            id=18,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR1'),
+        19:
+        dict(
+            name='forelegL2',
+            id=19,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR2'),
+        20:
+        dict(
+            name='forelegL3',
+            id=20,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR3'),
+        21:
+        dict(
+            name='forelegL4',
+            id=21,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR4'),
+        22:
+        dict(
+            name='midlegL1',
+            id=22,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR1'),
+        23:
+        dict(
+            name='midlegL2',
+            id=23,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR2'),
+        24:
+        dict(
+            name='midlegL3',
+            id=24,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR3'),
+        25:
+        dict(
+            name='midlegL4',
+            id=25,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR4'),
+        26:
+        dict(
+            name='hindlegL1',
+            id=26,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR1'),
+        27:
+        dict(
+            name='hindlegL2',
+            id=27,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR2'),
+        28:
+        dict(
+            name='hindlegL3',
+            id=28,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR3'),
+        29:
+        dict(
+            name='hindlegL4',
+            id=29,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR4'),
+        30:
+        dict(
+            name='wingL', id=30, color=[255, 255, 255], type='', swap='wingR'),
+        31:
+        dict(
+            name='wingR', id=31, color=[255, 255, 255], type='', swap='wingL'),
+    },
+    skeleton_info={
+        0: dict(link=('eyeL', 'head'), id=0, color=[255, 255, 255]),
+        1: dict(link=('eyeR', 'head'), id=1, color=[255, 255, 255]),
+        2: dict(link=('neck', 'head'), id=2, color=[255, 255, 255]),
+        3: dict(link=('thorax', 'neck'), id=3, color=[255, 255, 255]),
+        4: dict(link=('abdomen', 'thorax'), id=4, color=[255, 255, 255]),
+        5: dict(link=('forelegR2', 'forelegR1'), id=5, color=[255, 255, 255]),
+        6: dict(link=('forelegR3', 'forelegR2'), id=6, color=[255, 255, 255]),
+        7: dict(link=('forelegR4', 'forelegR3'), id=7, color=[255, 255, 255]),
+        8: dict(link=('midlegR2', 'midlegR1'), id=8, color=[255, 255, 255]),
+        9: dict(link=('midlegR3', 'midlegR2'), id=9, color=[255, 255, 255]),
+        10: dict(link=('midlegR4', 'midlegR3'), id=10, color=[255, 255, 255]),
+        11:
+        dict(link=('hindlegR2', 'hindlegR1'), id=11, color=[255, 255, 255]),
+        12:
+        dict(link=('hindlegR3', 'hindlegR2'), id=12, color=[255, 255, 255]),
+        13:
+        dict(link=('hindlegR4', 'hindlegR3'), id=13, color=[255, 255, 255]),
+        14:
+        dict(link=('forelegL2', 'forelegL1'), id=14, color=[255, 255, 255]),
+        15:
+        dict(link=('forelegL3', 'forelegL2'), id=15, color=[255, 255, 255]),
+        16:
+        dict(link=('forelegL4', 'forelegL3'), id=16, color=[255, 255, 255]),
+        17: dict(link=('midlegL2', 'midlegL1'), id=17, color=[255, 255, 255]),
+        18: dict(link=('midlegL3', 'midlegL2'), id=18, color=[255, 255, 255]),
+        19: dict(link=('midlegL4', 'midlegL3'), id=19, color=[255, 255, 255]),
+        20:
+        dict(link=('hindlegL2', 'hindlegL1'), id=20, color=[255, 255, 255]),
+        21:
+        dict(link=('hindlegL3', 'hindlegL2'), id=21, color=[255, 255, 255]),
+        22:
+        dict(link=('hindlegL4', 'hindlegL3'), id=22, color=[255, 255, 255]),
+        23: dict(link=('wingL', 'neck'), id=23, color=[255, 255, 255]),
+        24: dict(link=('wingR', 'neck'), id=24, color=[255, 255, 255])
+    },
+    joint_weights=[1.] * 32,
+    sigmas=[])

configs/_base_/datasets/freihand2d.py

@@ -0,0 +1,144 @@
+dataset_info = dict(
+    dataset_name='freihand',
+    paper_info=dict(
+        author='Zimmermann, Christian and Ceylan, Duygu and '
+        'Yang, Jimei and Russell, Bryan and '
+        'Argus, Max and Brox, Thomas',
+        title='Freihand: A dataset for markerless capture of hand pose '
+        'and shape from single rgb images',
+        container='Proceedings of the IEEE International '
+        'Conference on Computer Vision',
+        year='2019',
+        homepage='https://lmb.informatik.uni-freiburg.de/projects/freihand/',
+    ),
+    keypoint_info={
+        0:
+        dict(name='wrist', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='thumb1', id=1, color=[255, 128, 0], type='', swap=''),
+        2:
+        dict(name='thumb2', id=2, color=[255, 128, 0], type='', swap=''),
+        3:
+        dict(name='thumb3', id=3, color=[255, 128, 0], type='', swap=''),
+        4:
+        dict(name='thumb4', id=4, color=[255, 128, 0], type='', swap=''),
+        5:
+        dict(
+            name='forefinger1', id=5, color=[255, 153, 255], type='', swap=''),
+        6:
+        dict(
+            name='forefinger2', id=6, color=[255, 153, 255], type='', swap=''),
+        7:
+        dict(
+            name='forefinger3', id=7, color=[255, 153, 255], type='', swap=''),
+        8:
+        dict(
+            name='forefinger4', id=8, color=[255, 153, 255], type='', swap=''),
+        9:
+        dict(
+            name='middle_finger1',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        10:
+        dict(
+            name='middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        11:
+        dict(
+            name='middle_finger3',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        12:
+        dict(
+            name='middle_finger4',
+            id=12,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        13:
+        dict(
+            name='ring_finger1', id=13, color=[255, 51, 51], type='', swap=''),
+        14:
+        dict(
+            name='ring_finger2', id=14, color=[255, 51, 51], type='', swap=''),
+        15:
+        dict(
+            name='ring_finger3', id=15, color=[255, 51, 51], type='', swap=''),
+        16:
+        dict(
+            name='ring_finger4', id=16, color=[255, 51, 51], type='', swap=''),
+        17:
+        dict(name='pinky_finger1', id=17, color=[0, 255, 0], type='', swap=''),
+        18:
+        dict(name='pinky_finger2', id=18, color=[0, 255, 0], type='', swap=''),
+        19:
+        dict(name='pinky_finger3', id=19, color=[0, 255, 0], type='', swap=''),
+        20:
+        dict(name='pinky_finger4', id=20, color=[0, 255, 0], type='', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('wrist', 'thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('thumb1', 'thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('thumb2', 'thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('thumb3', 'thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('wrist', 'forefinger1'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('forefinger1', 'forefinger2'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('forefinger2', 'forefinger3'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('forefinger3', 'forefinger4'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('wrist', 'middle_finger1'), id=8, color=[102, 178, 255]),
+        9:
+        dict(
+            link=('middle_finger1', 'middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('middle_finger2', 'middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('middle_finger3', 'middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(link=('wrist', 'ring_finger1'), id=12, color=[255, 51, 51]),
+        13:
+        dict(
+            link=('ring_finger1', 'ring_finger2'), id=13, color=[255, 51, 51]),
+        14:
+        dict(
+            link=('ring_finger2', 'ring_finger3'), id=14, color=[255, 51, 51]),
+        15:
+        dict(
+            link=('ring_finger3', 'ring_finger4'), id=15, color=[255, 51, 51]),
+        16:
+        dict(link=('wrist', 'pinky_finger1'), id=16, color=[0, 255, 0]),
+        17:
+        dict(
+            link=('pinky_finger1', 'pinky_finger2'), id=17, color=[0, 255, 0]),
+        18:
+        dict(
+            link=('pinky_finger2', 'pinky_finger3'), id=18, color=[0, 255, 0]),
+        19:
+        dict(
+            link=('pinky_finger3', 'pinky_finger4'), id=19, color=[0, 255, 0])
+    },
+    joint_weights=[1.] * 21,
+    sigmas=[])

configs/_base_/datasets/h36m.py

@@ -0,0 +1,152 @@
+dataset_info = dict(
+    dataset_name='h36m',
+    paper_info=dict(
+        author='Ionescu, Catalin and Papava, Dragos and '
+        'Olaru, Vlad and Sminchisescu, Cristian',
+        title='Human3.6M: Large Scale Datasets and Predictive '
+        'Methods for 3D Human Sensing in Natural Environments',
+        container='IEEE Transactions on Pattern Analysis and '
+        'Machine Intelligence',
+        year='2014',
+        homepage='http://vision.imar.ro/human3.6m/description.php',
+    ),
+    keypoint_info={
+        0:
+        dict(name='root', id=0, color=[51, 153, 255], type='lower', swap=''),
+        1:
+        dict(
+            name='right_hip',
+            id=1,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        2:
+        dict(
+            name='right_knee',
+            id=2,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        3:
+        dict(
+            name='right_foot',
+            id=3,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_foot'),
+        4:
+        dict(
+            name='left_hip',
+            id=4,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        5:
+        dict(
+            name='left_knee',
+            id=5,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        6:
+        dict(
+            name='left_foot',
+            id=6,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_foot'),
+        7:
+        dict(name='spine', id=7, color=[51, 153, 255], type='upper', swap=''),
+        8:
+        dict(name='thorax', id=8, color=[51, 153, 255], type='upper', swap=''),
+        9:
+        dict(
+            name='neck_base',
+            id=9,
+            color=[51, 153, 255],
+            type='upper',
+            swap=''),
+        10:
+        dict(name='head', id=10, color=[51, 153, 255], type='upper', swap=''),
+        11:
+        dict(
+            name='left_shoulder',
+            id=11,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        12:
+        dict(
+            name='left_elbow',
+            id=12,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        13:
+        dict(
+            name='left_wrist',
+            id=13,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        14:
+        dict(
+            name='right_shoulder',
+            id=14,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        15:
+        dict(
+            name='right_elbow',
+            id=15,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        16:
+        dict(
+            name='right_wrist',
+            id=16,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist')
+    },
+    skeleton_info={
+        0:
+        dict(link=('root', 'left_hip'), id=0, color=[0, 255, 0]),
+        1:
+        dict(link=('left_hip', 'left_knee'), id=1, color=[0, 255, 0]),
+        2:
+        dict(link=('left_knee', 'left_foot'), id=2, color=[0, 255, 0]),
+        3:
+        dict(link=('root', 'right_hip'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('right_hip', 'right_knee'), id=4, color=[255, 128, 0]),
+        5:
+        dict(link=('right_knee', 'right_foot'), id=5, color=[255, 128, 0]),
+        6:
+        dict(link=('root', 'spine'), id=6, color=[51, 153, 255]),
+        7:
+        dict(link=('spine', 'thorax'), id=7, color=[51, 153, 255]),
+        8:
+        dict(link=('thorax', 'neck_base'), id=8, color=[51, 153, 255]),
+        9:
+        dict(link=('neck_base', 'head'), id=9, color=[51, 153, 255]),
+        10:
+        dict(link=('thorax', 'left_shoulder'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('left_shoulder', 'left_elbow'), id=11, color=[0, 255, 0]),
+        12:
+        dict(link=('left_elbow', 'left_wrist'), id=12, color=[0, 255, 0]),
+        13:
+        dict(link=('thorax', 'right_shoulder'), id=13, color=[255, 128, 0]),
+        14:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=14, color=[255, 128,
+                                                                  0]),
+        15:
+        dict(link=('right_elbow', 'right_wrist'), id=15, color=[255, 128, 0])
+    },
+    joint_weights=[1.] * 17,
+    sigmas=[],
+    stats_info=dict(bbox_center=(528., 427.), bbox_scale=400.))

configs/_base_/datasets/horse10.py

@@ -0,0 +1,201 @@
+dataset_info = dict(
+    dataset_name='horse10',
+    paper_info=dict(
+        author='Mathis, Alexander and Biasi, Thomas and '
+        'Schneider, Steffen and '
+        'Yuksekgonul, Mert and Rogers, Byron and '
+        'Bethge, Matthias and '
+        'Mathis, Mackenzie W',
+        title='Pretraining boosts out-of-domain robustness '
+        'for pose estimation',
+        container='Proceedings of the IEEE/CVF Winter Conference on '
+        'Applications of Computer Vision',
+        year='2021',
+        homepage='http://www.mackenziemathislab.org/horse10',
+    ),
+    keypoint_info={
+        0:
+        dict(name='Nose', id=0, color=[255, 153, 255], type='upper', swap=''),
+        1:
+        dict(name='Eye', id=1, color=[255, 153, 255], type='upper', swap=''),
+        2:
+        dict(
+            name='Nearknee',
+            id=2,
+            color=[255, 102, 255],
+            type='upper',
+            swap=''),
+        3:
+        dict(
+            name='Nearfrontfetlock',
+            id=3,
+            color=[255, 102, 255],
+            type='upper',
+            swap=''),
+        4:
+        dict(
+            name='Nearfrontfoot',
+            id=4,
+            color=[255, 102, 255],
+            type='upper',
+            swap=''),
+        5:
+        dict(
+            name='Offknee', id=5, color=[255, 102, 255], type='upper',
+            swap=''),
+        6:
+        dict(
+            name='Offfrontfetlock',
+            id=6,
+            color=[255, 102, 255],
+            type='upper',
+            swap=''),
+        7:
+        dict(
+            name='Offfrontfoot',
+            id=7,
+            color=[255, 102, 255],
+            type='upper',
+            swap=''),
+        8:
+        dict(
+            name='Shoulder',
+            id=8,
+            color=[255, 153, 255],
+            type='upper',
+            swap=''),
+        9:
+        dict(
+            name='Midshoulder',
+            id=9,
+            color=[255, 153, 255],
+            type='upper',
+            swap=''),
+        10:
+        dict(
+            name='Elbow', id=10, color=[255, 153, 255], type='upper', swap=''),
+        11:
+        dict(
+            name='Girth', id=11, color=[255, 153, 255], type='upper', swap=''),
+        12:
+        dict(
+            name='Wither', id=12, color=[255, 153, 255], type='upper',
+            swap=''),
+        13:
+        dict(
+            name='Nearhindhock',
+            id=13,
+            color=[255, 51, 255],
+            type='lower',
+            swap=''),
+        14:
+        dict(
+            name='Nearhindfetlock',
+            id=14,
+            color=[255, 51, 255],
+            type='lower',
+            swap=''),
+        15:
+        dict(
+            name='Nearhindfoot',
+            id=15,
+            color=[255, 51, 255],
+            type='lower',
+            swap=''),
+        16:
+        dict(name='Hip', id=16, color=[255, 153, 255], type='lower', swap=''),
+        17:
+        dict(
+            name='Stifle', id=17, color=[255, 153, 255], type='lower',
+            swap=''),
+        18:
+        dict(
+            name='Offhindhock',
+            id=18,
+            color=[255, 51, 255],
+            type='lower',
+            swap=''),
+        19:
+        dict(
+            name='Offhindfetlock',
+            id=19,
+            color=[255, 51, 255],
+            type='lower',
+            swap=''),
+        20:
+        dict(
+            name='Offhindfoot',
+            id=20,
+            color=[255, 51, 255],
+            type='lower',
+            swap=''),
+        21:
+        dict(
+            name='Ischium',
+            id=21,
+            color=[255, 153, 255],
+            type='lower',
+            swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('Nose', 'Eye'), id=0, color=[255, 153, 255]),
+        1:
+        dict(link=('Eye', 'Wither'), id=1, color=[255, 153, 255]),
+        2:
+        dict(link=('Wither', 'Hip'), id=2, color=[255, 153, 255]),
+        3:
+        dict(link=('Hip', 'Ischium'), id=3, color=[255, 153, 255]),
+        4:
+        dict(link=('Ischium', 'Stifle'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('Stifle', 'Girth'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('Girth', 'Elbow'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('Elbow', 'Shoulder'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('Shoulder', 'Midshoulder'), id=8, color=[255, 153, 255]),
+        9:
+        dict(link=('Midshoulder', 'Wither'), id=9, color=[255, 153, 255]),
+        10:
+        dict(
+            link=('Nearknee', 'Nearfrontfetlock'),
+            id=10,
+            color=[255, 102, 255]),
+        11:
+        dict(
+            link=('Nearfrontfetlock', 'Nearfrontfoot'),
+            id=11,
+            color=[255, 102, 255]),
+        12:
+        dict(
+            link=('Offknee', 'Offfrontfetlock'), id=12, color=[255, 102, 255]),
+        13:
+        dict(
+            link=('Offfrontfetlock', 'Offfrontfoot'),
+            id=13,
+            color=[255, 102, 255]),
+        14:
+        dict(
+            link=('Nearhindhock', 'Nearhindfetlock'),
+            id=14,
+            color=[255, 51, 255]),
+        15:
+        dict(
+            link=('Nearhindfetlock', 'Nearhindfoot'),
+            id=15,
+            color=[255, 51, 255]),
+        16:
+        dict(
+            link=('Offhindhock', 'Offhindfetlock'),
+            id=16,
+            color=[255, 51, 255]),
+        17:
+        dict(
+            link=('Offhindfetlock', 'Offhindfoot'),
+            id=17,
+            color=[255, 51, 255])
+    },
+    joint_weights=[1.] * 22,
+    sigmas=[])

configs/_base_/datasets/interhand2d.py

@@ -0,0 +1,142 @@
+dataset_info = dict(
+    dataset_name='interhand2d',
+    paper_info=dict(
+        author='Moon, Gyeongsik and Yu, Shoou-I and Wen, He and '
+        'Shiratori, Takaaki and Lee, Kyoung Mu',
+        title='InterHand2.6M: A dataset and baseline for 3D '
+        'interacting hand pose estimation from a single RGB image',
+        container='arXiv',
+        year='2020',
+        homepage='https://mks0601.github.io/InterHand2.6M/',
+    ),
+    keypoint_info={
+        0:
+        dict(name='thumb4', id=0, color=[255, 128, 0], type='', swap=''),
+        1:
+        dict(name='thumb3', id=1, color=[255, 128, 0], type='', swap=''),
+        2:
+        dict(name='thumb2', id=2, color=[255, 128, 0], type='', swap=''),
+        3:
+        dict(name='thumb1', id=3, color=[255, 128, 0], type='', swap=''),
+        4:
+        dict(
+            name='forefinger4', id=4, color=[255, 153, 255], type='', swap=''),
+        5:
+        dict(
+            name='forefinger3', id=5, color=[255, 153, 255], type='', swap=''),
+        6:
+        dict(
+            name='forefinger2', id=6, color=[255, 153, 255], type='', swap=''),
+        7:
+        dict(
+            name='forefinger1', id=7, color=[255, 153, 255], type='', swap=''),
+        8:
+        dict(
+            name='middle_finger4',
+            id=8,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        9:
+        dict(
+            name='middle_finger3',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        10:
+        dict(
+            name='middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        11:
+        dict(
+            name='middle_finger1',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        12:
+        dict(
+            name='ring_finger4', id=12, color=[255, 51, 51], type='', swap=''),
+        13:
+        dict(
+            name='ring_finger3', id=13, color=[255, 51, 51], type='', swap=''),
+        14:
+        dict(
+            name='ring_finger2', id=14, color=[255, 51, 51], type='', swap=''),
+        15:
+        dict(
+            name='ring_finger1', id=15, color=[255, 51, 51], type='', swap=''),
+        16:
+        dict(name='pinky_finger4', id=16, color=[0, 255, 0], type='', swap=''),
+        17:
+        dict(name='pinky_finger3', id=17, color=[0, 255, 0], type='', swap=''),
+        18:
+        dict(name='pinky_finger2', id=18, color=[0, 255, 0], type='', swap=''),
+        19:
+        dict(name='pinky_finger1', id=19, color=[0, 255, 0], type='', swap=''),
+        20:
+        dict(name='wrist', id=20, color=[255, 255, 255], type='', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('wrist', 'thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('thumb1', 'thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('thumb2', 'thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('thumb3', 'thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('wrist', 'forefinger1'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('forefinger1', 'forefinger2'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('forefinger2', 'forefinger3'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('forefinger3', 'forefinger4'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('wrist', 'middle_finger1'), id=8, color=[102, 178, 255]),
+        9:
+        dict(
+            link=('middle_finger1', 'middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('middle_finger2', 'middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('middle_finger3', 'middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(link=('wrist', 'ring_finger1'), id=12, color=[255, 51, 51]),
+        13:
+        dict(
+            link=('ring_finger1', 'ring_finger2'), id=13, color=[255, 51, 51]),
+        14:
+        dict(
+            link=('ring_finger2', 'ring_finger3'), id=14, color=[255, 51, 51]),
+        15:
+        dict(
+            link=('ring_finger3', 'ring_finger4'), id=15, color=[255, 51, 51]),
+        16:
+        dict(link=('wrist', 'pinky_finger1'), id=16, color=[0, 255, 0]),
+        17:
+        dict(
+            link=('pinky_finger1', 'pinky_finger2'), id=17, color=[0, 255, 0]),
+        18:
+        dict(
+            link=('pinky_finger2', 'pinky_finger3'), id=18, color=[0, 255, 0]),
+        19:
+        dict(
+            link=('pinky_finger3', 'pinky_finger4'), id=19, color=[0, 255, 0])
+    },
+    joint_weights=[1.] * 21,
+    sigmas=[])

configs/_base_/datasets/interhand3d.py

@@ -0,0 +1,487 @@
+dataset_info = dict(
+    dataset_name='interhand3d',
+    paper_info=dict(
+        author='Moon, Gyeongsik and Yu, Shoou-I and Wen, He and '
+        'Shiratori, Takaaki and Lee, Kyoung Mu',
+        title='InterHand2.6M: A dataset and baseline for 3D '
+        'interacting hand pose estimation from a single RGB image',
+        container='arXiv',
+        year='2020',
+        homepage='https://mks0601.github.io/InterHand2.6M/',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='right_thumb4',
+            id=0,
+            color=[255, 128, 0],
+            type='',
+            swap='left_thumb4'),
+        1:
+        dict(
+            name='right_thumb3',
+            id=1,
+            color=[255, 128, 0],
+            type='',
+            swap='left_thumb3'),
+        2:
+        dict(
+            name='right_thumb2',
+            id=2,
+            color=[255, 128, 0],
+            type='',
+            swap='left_thumb2'),
+        3:
+        dict(
+            name='right_thumb1',
+            id=3,
+            color=[255, 128, 0],
+            type='',
+            swap='left_thumb1'),
+        4:
+        dict(
+            name='right_forefinger4',
+            id=4,
+            color=[255, 153, 255],
+            type='',
+            swap='left_forefinger4'),
+        5:
+        dict(
+            name='right_forefinger3',
+            id=5,
+            color=[255, 153, 255],
+            type='',
+            swap='left_forefinger3'),
+        6:
+        dict(
+            name='right_forefinger2',
+            id=6,
+            color=[255, 153, 255],
+            type='',
+            swap='left_forefinger2'),
+        7:
+        dict(
+            name='right_forefinger1',
+            id=7,
+            color=[255, 153, 255],
+            type='',
+            swap='left_forefinger1'),
+        8:
+        dict(
+            name='right_middle_finger4',
+            id=8,
+            color=[102, 178, 255],
+            type='',
+            swap='left_middle_finger4'),
+        9:
+        dict(
+            name='right_middle_finger3',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap='left_middle_finger3'),
+        10:
+        dict(
+            name='right_middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap='left_middle_finger2'),
+        11:
+        dict(
+            name='right_middle_finger1',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap='left_middle_finger1'),
+        12:
+        dict(
+            name='right_ring_finger4',
+            id=12,
+            color=[255, 51, 51],
+            type='',
+            swap='left_ring_finger4'),
+        13:
+        dict(
+            name='right_ring_finger3',
+            id=13,
+            color=[255, 51, 51],
+            type='',
+            swap='left_ring_finger3'),
+        14:
+        dict(
+            name='right_ring_finger2',
+            id=14,
+            color=[255, 51, 51],
+            type='',
+            swap='left_ring_finger2'),
+        15:
+        dict(
+            name='right_ring_finger1',
+            id=15,
+            color=[255, 51, 51],
+            type='',
+            swap='left_ring_finger1'),
+        16:
+        dict(
+            name='right_pinky_finger4',
+            id=16,
+            color=[0, 255, 0],
+            type='',
+            swap='left_pinky_finger4'),
+        17:
+        dict(
+            name='right_pinky_finger3',
+            id=17,
+            color=[0, 255, 0],
+            type='',
+            swap='left_pinky_finger3'),
+        18:
+        dict(
+            name='right_pinky_finger2',
+            id=18,
+            color=[0, 255, 0],
+            type='',
+            swap='left_pinky_finger2'),
+        19:
+        dict(
+            name='right_pinky_finger1',
+            id=19,
+            color=[0, 255, 0],
+            type='',
+            swap='left_pinky_finger1'),
+        20:
+        dict(
+            name='right_wrist',
+            id=20,
+            color=[255, 255, 255],
+            type='',
+            swap='left_wrist'),
+        21:
+        dict(
+            name='left_thumb4',
+            id=21,
+            color=[255, 128, 0],
+            type='',
+            swap='right_thumb4'),
+        22:
+        dict(
+            name='left_thumb3',
+            id=22,
+            color=[255, 128, 0],
+            type='',
+            swap='right_thumb3'),
+        23:
+        dict(
+            name='left_thumb2',
+            id=23,
+            color=[255, 128, 0],
+            type='',
+            swap='right_thumb2'),
+        24:
+        dict(
+            name='left_thumb1',
+            id=24,
+            color=[255, 128, 0],
+            type='',
+            swap='right_thumb1'),
+        25:
+        dict(
+            name='left_forefinger4',
+            id=25,
+            color=[255, 153, 255],
+            type='',
+            swap='right_forefinger4'),
+        26:
+        dict(
+            name='left_forefinger3',
+            id=26,
+            color=[255, 153, 255],
+            type='',
+            swap='right_forefinger3'),
+        27:
+        dict(
+            name='left_forefinger2',
+            id=27,
+            color=[255, 153, 255],
+            type='',
+            swap='right_forefinger2'),
+        28:
+        dict(
+            name='left_forefinger1',
+            id=28,
+            color=[255, 153, 255],
+            type='',
+            swap='right_forefinger1'),
+        29:
+        dict(
+            name='left_middle_finger4',
+            id=29,
+            color=[102, 178, 255],
+            type='',
+            swap='right_middle_finger4'),
+        30:
+        dict(
+            name='left_middle_finger3',
+            id=30,
+            color=[102, 178, 255],
+            type='',
+            swap='right_middle_finger3'),
+        31:
+        dict(
+            name='left_middle_finger2',
+            id=31,
+            color=[102, 178, 255],
+            type='',
+            swap='right_middle_finger2'),
+        32:
+        dict(
+            name='left_middle_finger1',
+            id=32,
+            color=[102, 178, 255],
+            type='',
+            swap='right_middle_finger1'),
+        33:
+        dict(
+            name='left_ring_finger4',
+            id=33,
+            color=[255, 51, 51],
+            type='',
+            swap='right_ring_finger4'),
+        34:
+        dict(
+            name='left_ring_finger3',
+            id=34,
+            color=[255, 51, 51],
+            type='',
+            swap='right_ring_finger3'),
+        35:
+        dict(
+            name='left_ring_finger2',
+            id=35,
+            color=[255, 51, 51],
+            type='',
+            swap='right_ring_finger2'),
+        36:
+        dict(
+            name='left_ring_finger1',
+            id=36,
+            color=[255, 51, 51],
+            type='',
+            swap='right_ring_finger1'),
+        37:
+        dict(
+            name='left_pinky_finger4',
+            id=37,
+            color=[0, 255, 0],
+            type='',
+            swap='right_pinky_finger4'),
+        38:
+        dict(
+            name='left_pinky_finger3',
+            id=38,
+            color=[0, 255, 0],
+            type='',
+            swap='right_pinky_finger3'),
+        39:
+        dict(
+            name='left_pinky_finger2',
+            id=39,
+            color=[0, 255, 0],
+            type='',
+            swap='right_pinky_finger2'),
+        40:
+        dict(
+            name='left_pinky_finger1',
+            id=40,
+            color=[0, 255, 0],
+            type='',
+            swap='right_pinky_finger1'),
+        41:
+        dict(
+            name='left_wrist',
+            id=41,
+            color=[255, 255, 255],
+            type='',
+            swap='right_wrist'),
+    },
+    skeleton_info={
+        0:
+        dict(link=('right_wrist', 'right_thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('right_thumb1', 'right_thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('right_thumb2', 'right_thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('right_thumb3', 'right_thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(
+            link=('right_wrist', 'right_forefinger1'),
+            id=4,
+            color=[255, 153, 255]),
+        5:
+        dict(
+            link=('right_forefinger1', 'right_forefinger2'),
+            id=5,
+            color=[255, 153, 255]),
+        6:
+        dict(
+            link=('right_forefinger2', 'right_forefinger3'),
+            id=6,
+            color=[255, 153, 255]),
+        7:
+        dict(
+            link=('right_forefinger3', 'right_forefinger4'),
+            id=7,
+            color=[255, 153, 255]),
+        8:
+        dict(
+            link=('right_wrist', 'right_middle_finger1'),
+            id=8,
+            color=[102, 178, 255]),
+        9:
+        dict(
+            link=('right_middle_finger1', 'right_middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('right_middle_finger2', 'right_middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('right_middle_finger3', 'right_middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(
+            link=('right_wrist', 'right_ring_finger1'),
+            id=12,
+            color=[255, 51, 51]),
+        13:
+        dict(
+            link=('right_ring_finger1', 'right_ring_finger2'),
+            id=13,
+            color=[255, 51, 51]),
+        14:
+        dict(
+            link=('right_ring_finger2', 'right_ring_finger3'),
+            id=14,
+            color=[255, 51, 51]),
+        15:
+        dict(
+            link=('right_ring_finger3', 'right_ring_finger4'),
+            id=15,
+            color=[255, 51, 51]),
+        16:
+        dict(
+            link=('right_wrist', 'right_pinky_finger1'),
+            id=16,
+            color=[0, 255, 0]),
+        17:
+        dict(
+            link=('right_pinky_finger1', 'right_pinky_finger2'),
+            id=17,
+            color=[0, 255, 0]),
+        18:
+        dict(
+            link=('right_pinky_finger2', 'right_pinky_finger3'),
+            id=18,
+            color=[0, 255, 0]),
+        19:
+        dict(
+            link=('right_pinky_finger3', 'right_pinky_finger4'),
+            id=19,
+            color=[0, 255, 0]),
+        20:
+        dict(link=('left_wrist', 'left_thumb1'), id=20, color=[255, 128, 0]),
+        21:
+        dict(link=('left_thumb1', 'left_thumb2'), id=21, color=[255, 128, 0]),
+        22:
+        dict(link=('left_thumb2', 'left_thumb3'), id=22, color=[255, 128, 0]),
+        23:
+        dict(link=('left_thumb3', 'left_thumb4'), id=23, color=[255, 128, 0]),
+        24:
+        dict(
+            link=('left_wrist', 'left_forefinger1'),
+            id=24,
+            color=[255, 153, 255]),
+        25:
+        dict(
+            link=('left_forefinger1', 'left_forefinger2'),
+            id=25,
+            color=[255, 153, 255]),
+        26:
+        dict(
+            link=('left_forefinger2', 'left_forefinger3'),
+            id=26,
+            color=[255, 153, 255]),
+        27:
+        dict(
+            link=('left_forefinger3', 'left_forefinger4'),
+            id=27,
+            color=[255, 153, 255]),
+        28:
+        dict(
+            link=('left_wrist', 'left_middle_finger1'),
+            id=28,
+            color=[102, 178, 255]),
+        29:
+        dict(
+            link=('left_middle_finger1', 'left_middle_finger2'),
+            id=29,
+            color=[102, 178, 255]),
+        30:
+        dict(
+            link=('left_middle_finger2', 'left_middle_finger3'),
+            id=30,
+            color=[102, 178, 255]),
+        31:
+        dict(
+            link=('left_middle_finger3', 'left_middle_finger4'),
+            id=31,
+            color=[102, 178, 255]),
+        32:
+        dict(
+            link=('left_wrist', 'left_ring_finger1'),
+            id=32,
+            color=[255, 51, 51]),
+        33:
+        dict(
+            link=('left_ring_finger1', 'left_ring_finger2'),
+            id=33,
+            color=[255, 51, 51]),
+        34:
+        dict(
+            link=('left_ring_finger2', 'left_ring_finger3'),
+            id=34,
+            color=[255, 51, 51]),
+        35:
+        dict(
+            link=('left_ring_finger3', 'left_ring_finger4'),
+            id=35,
+            color=[255, 51, 51]),
+        36:
+        dict(
+            link=('left_wrist', 'left_pinky_finger1'),
+            id=36,
+            color=[0, 255, 0]),
+        37:
+        dict(
+            link=('left_pinky_finger1', 'left_pinky_finger2'),
+            id=37,
+            color=[0, 255, 0]),
+        38:
+        dict(
+            link=('left_pinky_finger2', 'left_pinky_finger3'),
+            id=38,
+            color=[0, 255, 0]),
+        39:
+        dict(
+            link=('left_pinky_finger3', 'left_pinky_finger4'),
+            id=39,
+            color=[0, 255, 0]),
+    },
+    joint_weights=[1.] * 42,
+    sigmas=[])

configs/_base_/datasets/jhmdb.py

@@ -0,0 +1,129 @@
+dataset_info = dict(
+    dataset_name='jhmdb',
+    paper_info=dict(
+        author='H. Jhuang and J. Gall and S. Zuffi and '
+        'C. Schmid and M. J. Black',
+        title='Towards understanding action recognition',
+        container='International Conf. on Computer Vision (ICCV)',
+        year='2013',
+        homepage='http://jhmdb.is.tue.mpg.de/dataset',
+    ),
+    keypoint_info={
+        0:
+        dict(name='neck', id=0, color=[255, 128, 0], type='upper', swap=''),
+        1:
+        dict(name='belly', id=1, color=[255, 128, 0], type='upper', swap=''),
+        2:
+        dict(name='head', id=2, color=[255, 128, 0], type='upper', swap=''),
+        3:
+        dict(
+            name='right_shoulder',
+            id=3,
+            color=[0, 255, 0],
+            type='upper',
+            swap='left_shoulder'),
+        4:
+        dict(
+            name='left_shoulder',
+            id=4,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        5:
+        dict(
+            name='right_hip',
+            id=5,
+            color=[0, 255, 0],
+            type='lower',
+            swap='left_hip'),
+        6:
+        dict(
+            name='left_hip',
+            id=6,
+            color=[51, 153, 255],
+            type='lower',
+            swap='right_hip'),
+        7:
+        dict(
+            name='right_elbow',
+            id=7,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_elbow'),
+        8:
+        dict(
+            name='left_elbow',
+            id=8,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_elbow'),
+        9:
+        dict(
+            name='right_knee',
+            id=9,
+            color=[51, 153, 255],
+            type='lower',
+            swap='left_knee'),
+        10:
+        dict(
+            name='left_knee',
+            id=10,
+            color=[255, 128, 0],
+            type='lower',
+            swap='right_knee'),
+        11:
+        dict(
+            name='right_wrist',
+            id=11,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        12:
+        dict(
+            name='left_wrist',
+            id=12,
+            color=[255, 128, 0],
+            type='upper',
+            swap='right_wrist'),
+        13:
+        dict(
+            name='right_ankle',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='left_ankle'),
+        14:
+        dict(
+            name='left_ankle',
+            id=14,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle')
+    },
+    skeleton_info={
+        0: dict(link=('right_ankle', 'right_knee'), id=0, color=[255, 128, 0]),
+        1: dict(link=('right_knee', 'right_hip'), id=1, color=[255, 128, 0]),
+        2: dict(link=('right_hip', 'belly'), id=2, color=[255, 128, 0]),
+        3: dict(link=('belly', 'left_hip'), id=3, color=[0, 255, 0]),
+        4: dict(link=('left_hip', 'left_knee'), id=4, color=[0, 255, 0]),
+        5: dict(link=('left_knee', 'left_ankle'), id=5, color=[0, 255, 0]),
+        6: dict(link=('belly', 'neck'), id=6, color=[51, 153, 255]),
+        7: dict(link=('neck', 'head'), id=7, color=[51, 153, 255]),
+        8: dict(link=('neck', 'right_shoulder'), id=8, color=[255, 128, 0]),
+        9: dict(
+            link=('right_shoulder', 'right_elbow'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('right_elbow', 'right_wrist'), id=10, color=[255, 128, 0]),
+        11: dict(link=('neck', 'left_shoulder'), id=11, color=[0, 255, 0]),
+        12:
+        dict(link=('left_shoulder', 'left_elbow'), id=12, color=[0, 255, 0]),
+        13: dict(link=('left_elbow', 'left_wrist'), id=13, color=[0, 255, 0])
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.2, 1.2, 1.5, 1.5, 1.5, 1.5
+    ],
+    # Adapted from COCO dataset.
+    sigmas=[
+        0.025, 0.107, 0.025, 0.079, 0.079, 0.107, 0.107, 0.072, 0.072, 0.087,
+        0.087, 0.062, 0.062, 0.089, 0.089
+    ])

configs/_base_/datasets/locust.py

@@ -0,0 +1,263 @@
+dataset_info = dict(
+    dataset_name='locust',
+    paper_info=dict(
+        author='Graving, Jacob M and Chae, Daniel and Naik, Hemal and '
+        'Li, Liang and Koger, Benjamin and Costelloe, Blair R and '
+        'Couzin, Iain D',
+        title='DeepPoseKit, a software toolkit for fast and robust '
+        'animal pose estimation using deep learning',
+        container='Elife',
+        year='2019',
+        homepage='https://github.com/jgraving/DeepPoseKit-Data',
+    ),
+    keypoint_info={
+        0:
+        dict(name='head', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='neck', id=1, color=[255, 255, 255], type='', swap=''),
+        2:
+        dict(name='thorax', id=2, color=[255, 255, 255], type='', swap=''),
+        3:
+        dict(name='abdomen1', id=3, color=[255, 255, 255], type='', swap=''),
+        4:
+        dict(name='abdomen2', id=4, color=[255, 255, 255], type='', swap=''),
+        5:
+        dict(
+            name='anttipL',
+            id=5,
+            color=[255, 255, 255],
+            type='',
+            swap='anttipR'),
+        6:
+        dict(
+            name='antbaseL',
+            id=6,
+            color=[255, 255, 255],
+            type='',
+            swap='antbaseR'),
+        7:
+        dict(name='eyeL', id=7, color=[255, 255, 255], type='', swap='eyeR'),
+        8:
+        dict(
+            name='forelegL1',
+            id=8,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR1'),
+        9:
+        dict(
+            name='forelegL2',
+            id=9,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR2'),
+        10:
+        dict(
+            name='forelegL3',
+            id=10,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR3'),
+        11:
+        dict(
+            name='forelegL4',
+            id=11,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR4'),
+        12:
+        dict(
+            name='midlegL1',
+            id=12,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR1'),
+        13:
+        dict(
+            name='midlegL2',
+            id=13,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR2'),
+        14:
+        dict(
+            name='midlegL3',
+            id=14,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR3'),
+        15:
+        dict(
+            name='midlegL4',
+            id=15,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegR4'),
+        16:
+        dict(
+            name='hindlegL1',
+            id=16,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR1'),
+        17:
+        dict(
+            name='hindlegL2',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR2'),
+        18:
+        dict(
+            name='hindlegL3',
+            id=18,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR3'),
+        19:
+        dict(
+            name='hindlegL4',
+            id=19,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR4'),
+        20:
+        dict(
+            name='anttipR',
+            id=20,
+            color=[255, 255, 255],
+            type='',
+            swap='anttipL'),
+        21:
+        dict(
+            name='antbaseR',
+            id=21,
+            color=[255, 255, 255],
+            type='',
+            swap='antbaseL'),
+        22:
+        dict(name='eyeR', id=22, color=[255, 255, 255], type='', swap='eyeL'),
+        23:
+        dict(
+            name='forelegR1',
+            id=23,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL1'),
+        24:
+        dict(
+            name='forelegR2',
+            id=24,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL2'),
+        25:
+        dict(
+            name='forelegR3',
+            id=25,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL3'),
+        26:
+        dict(
+            name='forelegR4',
+            id=26,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL4'),
+        27:
+        dict(
+            name='midlegR1',
+            id=27,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL1'),
+        28:
+        dict(
+            name='midlegR2',
+            id=28,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL2'),
+        29:
+        dict(
+            name='midlegR3',
+            id=29,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL3'),
+        30:
+        dict(
+            name='midlegR4',
+            id=30,
+            color=[255, 255, 255],
+            type='',
+            swap='midlegL4'),
+        31:
+        dict(
+            name='hindlegR1',
+            id=31,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL1'),
+        32:
+        dict(
+            name='hindlegR2',
+            id=32,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL2'),
+        33:
+        dict(
+            name='hindlegR3',
+            id=33,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL3'),
+        34:
+        dict(
+            name='hindlegR4',
+            id=34,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL4')
+    },
+    skeleton_info={
+        0: dict(link=('neck', 'head'), id=0, color=[255, 255, 255]),
+        1: dict(link=('thorax', 'neck'), id=1, color=[255, 255, 255]),
+        2: dict(link=('abdomen1', 'thorax'), id=2, color=[255, 255, 255]),
+        3: dict(link=('abdomen2', 'abdomen1'), id=3, color=[255, 255, 255]),
+        4: dict(link=('antbaseL', 'anttipL'), id=4, color=[255, 255, 255]),
+        5: dict(link=('eyeL', 'antbaseL'), id=5, color=[255, 255, 255]),
+        6: dict(link=('forelegL2', 'forelegL1'), id=6, color=[255, 255, 255]),
+        7: dict(link=('forelegL3', 'forelegL2'), id=7, color=[255, 255, 255]),
+        8: dict(link=('forelegL4', 'forelegL3'), id=8, color=[255, 255, 255]),
+        9: dict(link=('midlegL2', 'midlegL1'), id=9, color=[255, 255, 255]),
+        10: dict(link=('midlegL3', 'midlegL2'), id=10, color=[255, 255, 255]),
+        11: dict(link=('midlegL4', 'midlegL3'), id=11, color=[255, 255, 255]),
+        12:
+        dict(link=('hindlegL2', 'hindlegL1'), id=12, color=[255, 255, 255]),
+        13:
+        dict(link=('hindlegL3', 'hindlegL2'), id=13, color=[255, 255, 255]),
+        14:
+        dict(link=('hindlegL4', 'hindlegL3'), id=14, color=[255, 255, 255]),
+        15: dict(link=('antbaseR', 'anttipR'), id=15, color=[255, 255, 255]),
+        16: dict(link=('eyeR', 'antbaseR'), id=16, color=[255, 255, 255]),
+        17:
+        dict(link=('forelegR2', 'forelegR1'), id=17, color=[255, 255, 255]),
+        18:
+        dict(link=('forelegR3', 'forelegR2'), id=18, color=[255, 255, 255]),
+        19:
+        dict(link=('forelegR4', 'forelegR3'), id=19, color=[255, 255, 255]),
+        20: dict(link=('midlegR2', 'midlegR1'), id=20, color=[255, 255, 255]),
+        21: dict(link=('midlegR3', 'midlegR2'), id=21, color=[255, 255, 255]),
+        22: dict(link=('midlegR4', 'midlegR3'), id=22, color=[255, 255, 255]),
+        23:
+        dict(link=('hindlegR2', 'hindlegR1'), id=23, color=[255, 255, 255]),
+        24:
+        dict(link=('hindlegR3', 'hindlegR2'), id=24, color=[255, 255, 255]),
+        25:
+        dict(link=('hindlegR4', 'hindlegR3'), id=25, color=[255, 255, 255])
+    },
+    joint_weights=[1.] * 35,
+    sigmas=[])

configs/_base_/datasets/macaque.py

@@ -0,0 +1,183 @@
+dataset_info = dict(
+    dataset_name='macaque',
+    paper_info=dict(
+        author='Labuguen, Rollyn and Matsumoto, Jumpei and '
+        'Negrete, Salvador and Nishimaru, Hiroshi and '
+        'Nishijo, Hisao and Takada, Masahiko and '
+        'Go, Yasuhiro and Inoue, Ken-ichi and Shibata, Tomohiro',
+        title='MacaquePose: A novel "in the wild" macaque monkey pose dataset '
+        'for markerless motion capture',
+        container='bioRxiv',
+        year='2020',
+        homepage='http://www.pri.kyoto-u.ac.jp/datasets/'
+        'macaquepose/index.html',
+    ),
+    keypoint_info={
+        0:
+        dict(name='nose', id=0, color=[51, 153, 255], type='upper', swap=''),
+        1:
+        dict(
+            name='left_eye',
+            id=1,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_eye'),
+        2:
+        dict(
+            name='right_eye',
+            id=2,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_eye'),
+        3:
+        dict(
+            name='left_ear',
+            id=3,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_ear'),
+        4:
+        dict(
+            name='right_ear',
+            id=4,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_ear'),
+        5:
+        dict(
+            name='left_shoulder',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        6:
+        dict(
+            name='right_shoulder',
+            id=6,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        7:
+        dict(
+            name='left_elbow',
+            id=7,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        8:
+        dict(
+            name='right_elbow',
+            id=8,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        9:
+        dict(
+            name='left_wrist',
+            id=9,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        10:
+        dict(
+            name='right_wrist',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        11:
+        dict(
+            name='left_hip',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        12:
+        dict(
+            name='right_hip',
+            id=12,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        13:
+        dict(
+            name='left_knee',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        14:
+        dict(
+            name='right_knee',
+            id=14,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        15:
+        dict(
+            name='left_ankle',
+            id=15,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        16:
+        dict(
+            name='right_ankle',
+            id=16,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle')
+    },
+    skeleton_info={
+        0:
+        dict(link=('left_ankle', 'left_knee'), id=0, color=[0, 255, 0]),
+        1:
+        dict(link=('left_knee', 'left_hip'), id=1, color=[0, 255, 0]),
+        2:
+        dict(link=('right_ankle', 'right_knee'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('right_knee', 'right_hip'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('left_hip', 'right_hip'), id=4, color=[51, 153, 255]),
+        5:
+        dict(link=('left_shoulder', 'left_hip'), id=5, color=[51, 153, 255]),
+        6:
+        dict(link=('right_shoulder', 'right_hip'), id=6, color=[51, 153, 255]),
+        7:
+        dict(
+            link=('left_shoulder', 'right_shoulder'),
+            id=7,
+            color=[51, 153, 255]),
+        8:
+        dict(link=('left_shoulder', 'left_elbow'), id=8, color=[0, 255, 0]),
+        9:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('left_elbow', 'left_wrist'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('left_eye', 'right_eye'), id=12, color=[51, 153, 255]),
+        13:
+        dict(link=('nose', 'left_eye'), id=13, color=[51, 153, 255]),
+        14:
+        dict(link=('nose', 'right_eye'), id=14, color=[51, 153, 255]),
+        15:
+        dict(link=('left_eye', 'left_ear'), id=15, color=[51, 153, 255]),
+        16:
+        dict(link=('right_eye', 'right_ear'), id=16, color=[51, 153, 255]),
+        17:
+        dict(link=('left_ear', 'left_shoulder'), id=17, color=[51, 153, 255]),
+        18:
+        dict(
+            link=('right_ear', 'right_shoulder'), id=18, color=[51, 153, 255])
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.5, 1.5, 1., 1., 1.2, 1.2, 1.5,
+        1.5
+    ],
+    sigmas=[
+        0.026, 0.025, 0.025, 0.035, 0.035, 0.079, 0.079, 0.072, 0.072, 0.062,
+        0.062, 0.107, 0.107, 0.087, 0.087, 0.089, 0.089
+    ])

configs/_base_/datasets/mhp.py

@@ -0,0 +1,156 @@
+dataset_info = dict(
+    dataset_name='mhp',
+    paper_info=dict(
+        author='Zhao, Jian and Li, Jianshu and Cheng, Yu and '
+        'Sim, Terence and Yan, Shuicheng and Feng, Jiashi',
+        title='Understanding humans in crowded scenes: '
+        'Deep nested adversarial learning and a '
+        'new benchmark for multi-human parsing',
+        container='Proceedings of the 26th ACM '
+        'international conference on Multimedia',
+        year='2018',
+        homepage='https://lv-mhp.github.io/dataset',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='right_ankle',
+            id=0,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle'),
+        1:
+        dict(
+            name='right_knee',
+            id=1,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        2:
+        dict(
+            name='right_hip',
+            id=2,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        3:
+        dict(
+            name='left_hip',
+            id=3,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        4:
+        dict(
+            name='left_knee',
+            id=4,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        5:
+        dict(
+            name='left_ankle',
+            id=5,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        6:
+        dict(name='pelvis', id=6, color=[51, 153, 255], type='lower', swap=''),
+        7:
+        dict(name='thorax', id=7, color=[51, 153, 255], type='upper', swap=''),
+        8:
+        dict(
+            name='upper_neck',
+            id=8,
+            color=[51, 153, 255],
+            type='upper',
+            swap=''),
+        9:
+        dict(
+            name='head_top', id=9, color=[51, 153, 255], type='upper',
+            swap=''),
+        10:
+        dict(
+            name='right_wrist',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        11:
+        dict(
+            name='right_elbow',
+            id=11,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        12:
+        dict(
+            name='right_shoulder',
+            id=12,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        13:
+        dict(
+            name='left_shoulder',
+            id=13,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        14:
+        dict(
+            name='left_elbow',
+            id=14,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        15:
+        dict(
+            name='left_wrist',
+            id=15,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist')
+    },
+    skeleton_info={
+        0:
+        dict(link=('right_ankle', 'right_knee'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('right_knee', 'right_hip'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('right_hip', 'pelvis'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('pelvis', 'left_hip'), id=3, color=[0, 255, 0]),
+        4:
+        dict(link=('left_hip', 'left_knee'), id=4, color=[0, 255, 0]),
+        5:
+        dict(link=('left_knee', 'left_ankle'), id=5, color=[0, 255, 0]),
+        6:
+        dict(link=('pelvis', 'thorax'), id=6, color=[51, 153, 255]),
+        7:
+        dict(link=('thorax', 'upper_neck'), id=7, color=[51, 153, 255]),
+        8:
+        dict(link=('upper_neck', 'head_top'), id=8, color=[51, 153, 255]),
+        9:
+        dict(link=('upper_neck', 'right_shoulder'), id=9, color=[255, 128, 0]),
+        10:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=10, color=[255, 128,
+                                                                  0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('upper_neck', 'left_shoulder'), id=12, color=[0, 255, 0]),
+        13:
+        dict(link=('left_shoulder', 'left_elbow'), id=13, color=[0, 255, 0]),
+        14:
+        dict(link=('left_elbow', 'left_wrist'), id=14, color=[0, 255, 0])
+    },
+    joint_weights=[
+        1.5, 1.2, 1., 1., 1.2, 1.5, 1., 1., 1., 1., 1.5, 1.2, 1., 1., 1.2, 1.5
+    ],
+    # Adapted from COCO dataset.
+    sigmas=[
+        0.089, 0.083, 0.107, 0.107, 0.083, 0.089, 0.026, 0.026, 0.026, 0.026,
+        0.062, 0.072, 0.179, 0.179, 0.072, 0.062
+    ])

configs/_base_/datasets/mpi_inf_3dhp.py

@@ -0,0 +1,132 @@
+dataset_info = dict(
+    dataset_name='mpi_inf_3dhp',
+    paper_info=dict(
+        author='ehta, Dushyant and Rhodin, Helge and Casas, Dan and '
+        'Fua, Pascal and Sotnychenko, Oleksandr and Xu, Weipeng and '
+        'Theobalt, Christian',
+        title='Monocular 3D Human Pose Estimation In The Wild Using Improved '
+        'CNN Supervision',
+        container='2017 international conference on 3D vision (3DV)',
+        year='2017',
+        homepage='http://gvv.mpi-inf.mpg.de/3dhp-dataset',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='head_top', id=0, color=[51, 153, 255], type='upper',
+            swap=''),
+        1:
+        dict(name='neck', id=1, color=[51, 153, 255], type='upper', swap=''),
+        2:
+        dict(
+            name='right_shoulder',
+            id=2,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        3:
+        dict(
+            name='right_elbow',
+            id=3,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        4:
+        dict(
+            name='right_wrist',
+            id=4,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        5:
+        dict(
+            name='left_shoulder',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        6:
+        dict(
+            name='left_elbow',
+            id=6,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        7:
+        dict(
+            name='left_wrist',
+            id=7,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        8:
+        dict(
+            name='right_hip',
+            id=8,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        9:
+        dict(
+            name='right_knee',
+            id=9,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        10:
+        dict(
+            name='right_ankle',
+            id=10,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle'),
+        11:
+        dict(
+            name='left_hip',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        12:
+        dict(
+            name='left_knee',
+            id=12,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        13:
+        dict(
+            name='left_ankle',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        14:
+        dict(name='root', id=14, color=[51, 153, 255], type='lower', swap=''),
+        15:
+        dict(name='spine', id=15, color=[51, 153, 255], type='upper', swap=''),
+        16:
+        dict(name='head', id=16, color=[51, 153, 255], type='upper', swap='')
+    },
+    skeleton_info={
+        0: dict(link=('neck', 'right_shoulder'), id=0, color=[255, 128, 0]),
+        1: dict(
+            link=('right_shoulder', 'right_elbow'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('right_elbow', 'right_wrist'), id=2, color=[255, 128, 0]),
+        3: dict(link=('neck', 'left_shoulder'), id=3, color=[0, 255, 0]),
+        4: dict(link=('left_shoulder', 'left_elbow'), id=4, color=[0, 255, 0]),
+        5: dict(link=('left_elbow', 'left_wrist'), id=5, color=[0, 255, 0]),
+        6: dict(link=('root', 'right_hip'), id=6, color=[255, 128, 0]),
+        7: dict(link=('right_hip', 'right_knee'), id=7, color=[255, 128, 0]),
+        8: dict(link=('right_knee', 'right_ankle'), id=8, color=[255, 128, 0]),
+        9: dict(link=('root', 'left_hip'), id=9, color=[0, 255, 0]),
+        10: dict(link=('left_hip', 'left_knee'), id=10, color=[0, 255, 0]),
+        11: dict(link=('left_knee', 'left_ankle'), id=11, color=[0, 255, 0]),
+        12: dict(link=('head_top', 'head'), id=12, color=[51, 153, 255]),
+        13: dict(link=('head', 'neck'), id=13, color=[51, 153, 255]),
+        14: dict(link=('neck', 'spine'), id=14, color=[51, 153, 255]),
+        15: dict(link=('spine', 'root'), id=15, color=[51, 153, 255])
+    },
+    joint_weights=[1.] * 17,
+    sigmas=[])

configs/_base_/datasets/mpii.py

@@ -0,0 +1,155 @@
+dataset_info = dict(
+    dataset_name='mpii',
+    paper_info=dict(
+        author='Mykhaylo Andriluka and Leonid Pishchulin and '
+        'Peter Gehler and Schiele, Bernt',
+        title='2D Human Pose Estimation: New Benchmark and '
+        'State of the Art Analysis',
+        container='IEEE Conference on Computer Vision and '
+        'Pattern Recognition (CVPR)',
+        year='2014',
+        homepage='http://human-pose.mpi-inf.mpg.de/',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='right_ankle',
+            id=0,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle'),
+        1:
+        dict(
+            name='right_knee',
+            id=1,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        2:
+        dict(
+            name='right_hip',
+            id=2,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        3:
+        dict(
+            name='left_hip',
+            id=3,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        4:
+        dict(
+            name='left_knee',
+            id=4,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        5:
+        dict(
+            name='left_ankle',
+            id=5,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        6:
+        dict(name='pelvis', id=6, color=[51, 153, 255], type='lower', swap=''),
+        7:
+        dict(name='thorax', id=7, color=[51, 153, 255], type='upper', swap=''),
+        8:
+        dict(
+            name='upper_neck',
+            id=8,
+            color=[51, 153, 255],
+            type='upper',
+            swap=''),
+        9:
+        dict(
+            name='head_top', id=9, color=[51, 153, 255], type='upper',
+            swap=''),
+        10:
+        dict(
+            name='right_wrist',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        11:
+        dict(
+            name='right_elbow',
+            id=11,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        12:
+        dict(
+            name='right_shoulder',
+            id=12,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        13:
+        dict(
+            name='left_shoulder',
+            id=13,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        14:
+        dict(
+            name='left_elbow',
+            id=14,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        15:
+        dict(
+            name='left_wrist',
+            id=15,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist')
+    },
+    skeleton_info={
+        0:
+        dict(link=('right_ankle', 'right_knee'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('right_knee', 'right_hip'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('right_hip', 'pelvis'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('pelvis', 'left_hip'), id=3, color=[0, 255, 0]),
+        4:
+        dict(link=('left_hip', 'left_knee'), id=4, color=[0, 255, 0]),
+        5:
+        dict(link=('left_knee', 'left_ankle'), id=5, color=[0, 255, 0]),
+        6:
+        dict(link=('pelvis', 'thorax'), id=6, color=[51, 153, 255]),
+        7:
+        dict(link=('thorax', 'upper_neck'), id=7, color=[51, 153, 255]),
+        8:
+        dict(link=('upper_neck', 'head_top'), id=8, color=[51, 153, 255]),
+        9:
+        dict(link=('upper_neck', 'right_shoulder'), id=9, color=[255, 128, 0]),
+        10:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=10, color=[255, 128,
+                                                                  0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('upper_neck', 'left_shoulder'), id=12, color=[0, 255, 0]),
+        13:
+        dict(link=('left_shoulder', 'left_elbow'), id=13, color=[0, 255, 0]),
+        14:
+        dict(link=('left_elbow', 'left_wrist'), id=14, color=[0, 255, 0])
+    },
+    joint_weights=[
+        1.5, 1.2, 1., 1., 1.2, 1.5, 1., 1., 1., 1., 1.5, 1.2, 1., 1., 1.2, 1.5
+    ],
+    # Adapted from COCO dataset.
+    sigmas=[
+        0.089, 0.083, 0.107, 0.107, 0.083, 0.089, 0.026, 0.026, 0.026, 0.026,
+        0.062, 0.072, 0.179, 0.179, 0.072, 0.062
+    ])

configs/_base_/datasets/mpii_trb.py

@@ -0,0 +1,380 @@
+dataset_info = dict(
+    dataset_name='mpii_trb',
+    paper_info=dict(
+        author='Duan, Haodong and Lin, Kwan-Yee and Jin, Sheng and '
+        'Liu, Wentao and Qian, Chen and Ouyang, Wanli',
+        title='TRB: A Novel Triplet Representation for '
+        'Understanding 2D Human Body',
+        container='Proceedings of the IEEE International '
+        'Conference on Computer Vision',
+        year='2019',
+        homepage='https://github.com/kennymckormick/'
+        'Triplet-Representation-of-human-Body',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='left_shoulder',
+            id=0,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        1:
+        dict(
+            name='right_shoulder',
+            id=1,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        2:
+        dict(
+            name='left_elbow',
+            id=2,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        3:
+        dict(
+            name='right_elbow',
+            id=3,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        4:
+        dict(
+            name='left_wrist',
+            id=4,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        5:
+        dict(
+            name='right_wrist',
+            id=5,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        6:
+        dict(
+            name='left_hip',
+            id=6,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        7:
+        dict(
+            name='right_hip',
+            id=7,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        8:
+        dict(
+            name='left_knee',
+            id=8,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        9:
+        dict(
+            name='right_knee',
+            id=9,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        10:
+        dict(
+            name='left_ankle',
+            id=10,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        11:
+        dict(
+            name='right_ankle',
+            id=11,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle'),
+        12:
+        dict(name='head', id=12, color=[51, 153, 255], type='upper', swap=''),
+        13:
+        dict(name='neck', id=13, color=[51, 153, 255], type='upper', swap=''),
+        14:
+        dict(
+            name='right_neck',
+            id=14,
+            color=[255, 255, 255],
+            type='upper',
+            swap='left_neck'),
+        15:
+        dict(
+            name='left_neck',
+            id=15,
+            color=[255, 255, 255],
+            type='upper',
+            swap='right_neck'),
+        16:
+        dict(
+            name='medial_right_shoulder',
+            id=16,
+            color=[255, 255, 255],
+            type='upper',
+            swap='medial_left_shoulder'),
+        17:
+        dict(
+            name='lateral_right_shoulder',
+            id=17,
+            color=[255, 255, 255],
+            type='upper',
+            swap='lateral_left_shoulder'),
+        18:
+        dict(
+            name='medial_right_bow',
+            id=18,
+            color=[255, 255, 255],
+            type='upper',
+            swap='medial_left_bow'),
+        19:
+        dict(
+            name='lateral_right_bow',
+            id=19,
+            color=[255, 255, 255],
+            type='upper',
+            swap='lateral_left_bow'),
+        20:
+        dict(
+            name='medial_right_wrist',
+            id=20,
+            color=[255, 255, 255],
+            type='upper',
+            swap='medial_left_wrist'),
+        21:
+        dict(
+            name='lateral_right_wrist',
+            id=21,
+            color=[255, 255, 255],
+            type='upper',
+            swap='lateral_left_wrist'),
+        22:
+        dict(
+            name='medial_left_shoulder',
+            id=22,
+            color=[255, 255, 255],
+            type='upper',
+            swap='medial_right_shoulder'),
+        23:
+        dict(
+            name='lateral_left_shoulder',
+            id=23,
+            color=[255, 255, 255],
+            type='upper',
+            swap='lateral_right_shoulder'),
+        24:
+        dict(
+            name='medial_left_bow',
+            id=24,
+            color=[255, 255, 255],
+            type='upper',
+            swap='medial_right_bow'),
+        25:
+        dict(
+            name='lateral_left_bow',
+            id=25,
+            color=[255, 255, 255],
+            type='upper',
+            swap='lateral_right_bow'),
+        26:
+        dict(
+            name='medial_left_wrist',
+            id=26,
+            color=[255, 255, 255],
+            type='upper',
+            swap='medial_right_wrist'),
+        27:
+        dict(
+            name='lateral_left_wrist',
+            id=27,
+            color=[255, 255, 255],
+            type='upper',
+            swap='lateral_right_wrist'),
+        28:
+        dict(
+            name='medial_right_hip',
+            id=28,
+            color=[255, 255, 255],
+            type='lower',
+            swap='medial_left_hip'),
+        29:
+        dict(
+            name='lateral_right_hip',
+            id=29,
+            color=[255, 255, 255],
+            type='lower',
+            swap='lateral_left_hip'),
+        30:
+        dict(
+            name='medial_right_knee',
+            id=30,
+            color=[255, 255, 255],
+            type='lower',
+            swap='medial_left_knee'),
+        31:
+        dict(
+            name='lateral_right_knee',
+            id=31,
+            color=[255, 255, 255],
+            type='lower',
+            swap='lateral_left_knee'),
+        32:
+        dict(
+            name='medial_right_ankle',
+            id=32,
+            color=[255, 255, 255],
+            type='lower',
+            swap='medial_left_ankle'),
+        33:
+        dict(
+            name='lateral_right_ankle',
+            id=33,
+            color=[255, 255, 255],
+            type='lower',
+            swap='lateral_left_ankle'),
+        34:
+        dict(
+            name='medial_left_hip',
+            id=34,
+            color=[255, 255, 255],
+            type='lower',
+            swap='medial_right_hip'),
+        35:
+        dict(
+            name='lateral_left_hip',
+            id=35,
+            color=[255, 255, 255],
+            type='lower',
+            swap='lateral_right_hip'),
+        36:
+        dict(
+            name='medial_left_knee',
+            id=36,
+            color=[255, 255, 255],
+            type='lower',
+            swap='medial_right_knee'),
+        37:
+        dict(
+            name='lateral_left_knee',
+            id=37,
+            color=[255, 255, 255],
+            type='lower',
+            swap='lateral_right_knee'),
+        38:
+        dict(
+            name='medial_left_ankle',
+            id=38,
+            color=[255, 255, 255],
+            type='lower',
+            swap='medial_right_ankle'),
+        39:
+        dict(
+            name='lateral_left_ankle',
+            id=39,
+            color=[255, 255, 255],
+            type='lower',
+            swap='lateral_right_ankle'),
+    },
+    skeleton_info={
+        0:
+        dict(link=('head', 'neck'), id=0, color=[51, 153, 255]),
+        1:
+        dict(link=('neck', 'left_shoulder'), id=1, color=[51, 153, 255]),
+        2:
+        dict(link=('neck', 'right_shoulder'), id=2, color=[51, 153, 255]),
+        3:
+        dict(link=('left_shoulder', 'left_elbow'), id=3, color=[0, 255, 0]),
+        4:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=4, color=[255, 128, 0]),
+        5:
+        dict(link=('left_elbow', 'left_wrist'), id=5, color=[0, 255, 0]),
+        6:
+        dict(link=('right_elbow', 'right_wrist'), id=6, color=[255, 128, 0]),
+        7:
+        dict(link=('left_shoulder', 'left_hip'), id=7, color=[51, 153, 255]),
+        8:
+        dict(link=('right_shoulder', 'right_hip'), id=8, color=[51, 153, 255]),
+        9:
+        dict(link=('left_hip', 'right_hip'), id=9, color=[51, 153, 255]),
+        10:
+        dict(link=('left_hip', 'left_knee'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('right_hip', 'right_knee'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('left_knee', 'left_ankle'), id=12, color=[0, 255, 0]),
+        13:
+        dict(link=('right_knee', 'right_ankle'), id=13, color=[255, 128, 0]),
+        14:
+        dict(link=('right_neck', 'left_neck'), id=14, color=[255, 255, 255]),
+        15:
+        dict(
+            link=('medial_right_shoulder', 'lateral_right_shoulder'),
+            id=15,
+            color=[255, 255, 255]),
+        16:
+        dict(
+            link=('medial_right_bow', 'lateral_right_bow'),
+            id=16,
+            color=[255, 255, 255]),
+        17:
+        dict(
+            link=('medial_right_wrist', 'lateral_right_wrist'),
+            id=17,
+            color=[255, 255, 255]),
+        18:
+        dict(
+            link=('medial_left_shoulder', 'lateral_left_shoulder'),
+            id=18,
+            color=[255, 255, 255]),
+        19:
+        dict(
+            link=('medial_left_bow', 'lateral_left_bow'),
+            id=19,
+            color=[255, 255, 255]),
+        20:
+        dict(
+            link=('medial_left_wrist', 'lateral_left_wrist'),
+            id=20,
+            color=[255, 255, 255]),
+        21:
+        dict(
+            link=('medial_right_hip', 'lateral_right_hip'),
+            id=21,
+            color=[255, 255, 255]),
+        22:
+        dict(
+            link=('medial_right_knee', 'lateral_right_knee'),
+            id=22,
+            color=[255, 255, 255]),
+        23:
+        dict(
+            link=('medial_right_ankle', 'lateral_right_ankle'),
+            id=23,
+            color=[255, 255, 255]),
+        24:
+        dict(
+            link=('medial_left_hip', 'lateral_left_hip'),
+            id=24,
+            color=[255, 255, 255]),
+        25:
+        dict(
+            link=('medial_left_knee', 'lateral_left_knee'),
+            id=25,
+            color=[255, 255, 255]),
+        26:
+        dict(
+            link=('medial_left_ankle', 'lateral_left_ankle'),
+            id=26,
+            color=[255, 255, 255])
+    },
+    joint_weights=[1.] * 40,
+    sigmas=[])

configs/_base_/datasets/ochuman.py

@@ -0,0 +1,181 @@
+dataset_info = dict(
+    dataset_name='ochuman',
+    paper_info=dict(
+        author='Zhang, Song-Hai and Li, Ruilong and Dong, Xin and '
+        'Rosin, Paul and Cai, Zixi and Han, Xi and '
+        'Yang, Dingcheng and Huang, Haozhi and Hu, Shi-Min',
+        title='Pose2seg: Detection free human instance segmentation',
+        container='Proceedings of the IEEE conference on computer '
+        'vision and pattern recognition',
+        year='2019',
+        homepage='https://github.com/liruilong940607/OCHumanApi',
+    ),
+    keypoint_info={
+        0:
+        dict(name='nose', id=0, color=[51, 153, 255], type='upper', swap=''),
+        1:
+        dict(
+            name='left_eye',
+            id=1,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_eye'),
+        2:
+        dict(
+            name='right_eye',
+            id=2,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_eye'),
+        3:
+        dict(
+            name='left_ear',
+            id=3,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_ear'),
+        4:
+        dict(
+            name='right_ear',
+            id=4,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_ear'),
+        5:
+        dict(
+            name='left_shoulder',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        6:
+        dict(
+            name='right_shoulder',
+            id=6,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        7:
+        dict(
+            name='left_elbow',
+            id=7,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        8:
+        dict(
+            name='right_elbow',
+            id=8,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        9:
+        dict(
+            name='left_wrist',
+            id=9,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        10:
+        dict(
+            name='right_wrist',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        11:
+        dict(
+            name='left_hip',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        12:
+        dict(
+            name='right_hip',
+            id=12,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        13:
+        dict(
+            name='left_knee',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        14:
+        dict(
+            name='right_knee',
+            id=14,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        15:
+        dict(
+            name='left_ankle',
+            id=15,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        16:
+        dict(
+            name='right_ankle',
+            id=16,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle')
+    },
+    skeleton_info={
+        0:
+        dict(link=('left_ankle', 'left_knee'), id=0, color=[0, 255, 0]),
+        1:
+        dict(link=('left_knee', 'left_hip'), id=1, color=[0, 255, 0]),
+        2:
+        dict(link=('right_ankle', 'right_knee'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('right_knee', 'right_hip'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('left_hip', 'right_hip'), id=4, color=[51, 153, 255]),
+        5:
+        dict(link=('left_shoulder', 'left_hip'), id=5, color=[51, 153, 255]),
+        6:
+        dict(link=('right_shoulder', 'right_hip'), id=6, color=[51, 153, 255]),
+        7:
+        dict(
+            link=('left_shoulder', 'right_shoulder'),
+            id=7,
+            color=[51, 153, 255]),
+        8:
+        dict(link=('left_shoulder', 'left_elbow'), id=8, color=[0, 255, 0]),
+        9:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('left_elbow', 'left_wrist'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('left_eye', 'right_eye'), id=12, color=[51, 153, 255]),
+        13:
+        dict(link=('nose', 'left_eye'), id=13, color=[51, 153, 255]),
+        14:
+        dict(link=('nose', 'right_eye'), id=14, color=[51, 153, 255]),
+        15:
+        dict(link=('left_eye', 'left_ear'), id=15, color=[51, 153, 255]),
+        16:
+        dict(link=('right_eye', 'right_ear'), id=16, color=[51, 153, 255]),
+        17:
+        dict(link=('left_ear', 'left_shoulder'), id=17, color=[51, 153, 255]),
+        18:
+        dict(
+            link=('right_ear', 'right_shoulder'), id=18, color=[51, 153, 255])
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.5, 1.5, 1., 1., 1.2, 1.2, 1.5,
+        1.5
+    ],
+    sigmas=[
+        0.026, 0.025, 0.025, 0.035, 0.035, 0.079, 0.079, 0.072, 0.072, 0.062,
+        0.062, 0.107, 0.107, 0.087, 0.087, 0.089, 0.089
+    ])

configs/_base_/datasets/onehand10k.py

@@ -0,0 +1,142 @@
+dataset_info = dict(
+    dataset_name='onehand10k',
+    paper_info=dict(
+        author='Wang, Yangang and Peng, Cong and Liu, Yebin',
+        title='Mask-pose cascaded cnn for 2d hand pose estimation '
+        'from single color image',
+        container='IEEE Transactions on Circuits and Systems '
+        'for Video Technology',
+        year='2018',
+        homepage='https://www.yangangwang.com/papers/WANG-MCC-2018-10.html',
+    ),
+    keypoint_info={
+        0:
+        dict(name='wrist', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='thumb1', id=1, color=[255, 128, 0], type='', swap=''),
+        2:
+        dict(name='thumb2', id=2, color=[255, 128, 0], type='', swap=''),
+        3:
+        dict(name='thumb3', id=3, color=[255, 128, 0], type='', swap=''),
+        4:
+        dict(name='thumb4', id=4, color=[255, 128, 0], type='', swap=''),
+        5:
+        dict(
+            name='forefinger1', id=5, color=[255, 153, 255], type='', swap=''),
+        6:
+        dict(
+            name='forefinger2', id=6, color=[255, 153, 255], type='', swap=''),
+        7:
+        dict(
+            name='forefinger3', id=7, color=[255, 153, 255], type='', swap=''),
+        8:
+        dict(
+            name='forefinger4', id=8, color=[255, 153, 255], type='', swap=''),
+        9:
+        dict(
+            name='middle_finger1',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        10:
+        dict(
+            name='middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        11:
+        dict(
+            name='middle_finger3',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        12:
+        dict(
+            name='middle_finger4',
+            id=12,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        13:
+        dict(
+            name='ring_finger1', id=13, color=[255, 51, 51], type='', swap=''),
+        14:
+        dict(
+            name='ring_finger2', id=14, color=[255, 51, 51], type='', swap=''),
+        15:
+        dict(
+            name='ring_finger3', id=15, color=[255, 51, 51], type='', swap=''),
+        16:
+        dict(
+            name='ring_finger4', id=16, color=[255, 51, 51], type='', swap=''),
+        17:
+        dict(name='pinky_finger1', id=17, color=[0, 255, 0], type='', swap=''),
+        18:
+        dict(name='pinky_finger2', id=18, color=[0, 255, 0], type='', swap=''),
+        19:
+        dict(name='pinky_finger3', id=19, color=[0, 255, 0], type='', swap=''),
+        20:
+        dict(name='pinky_finger4', id=20, color=[0, 255, 0], type='', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('wrist', 'thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('thumb1', 'thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('thumb2', 'thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('thumb3', 'thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('wrist', 'forefinger1'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('forefinger1', 'forefinger2'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('forefinger2', 'forefinger3'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('forefinger3', 'forefinger4'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('wrist', 'middle_finger1'), id=8, color=[102, 178, 255]),
+        9:
+        dict(
+            link=('middle_finger1', 'middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('middle_finger2', 'middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('middle_finger3', 'middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(link=('wrist', 'ring_finger1'), id=12, color=[255, 51, 51]),
+        13:
+        dict(
+            link=('ring_finger1', 'ring_finger2'), id=13, color=[255, 51, 51]),
+        14:
+        dict(
+            link=('ring_finger2', 'ring_finger3'), id=14, color=[255, 51, 51]),
+        15:
+        dict(
+            link=('ring_finger3', 'ring_finger4'), id=15, color=[255, 51, 51]),
+        16:
+        dict(link=('wrist', 'pinky_finger1'), id=16, color=[0, 255, 0]),
+        17:
+        dict(
+            link=('pinky_finger1', 'pinky_finger2'), id=17, color=[0, 255, 0]),
+        18:
+        dict(
+            link=('pinky_finger2', 'pinky_finger3'), id=18, color=[0, 255, 0]),
+        19:
+        dict(
+            link=('pinky_finger3', 'pinky_finger4'), id=19, color=[0, 255, 0])
+    },
+    joint_weights=[1.] * 21,
+    sigmas=[])

configs/_base_/datasets/panoptic_body3d.py

@@ -0,0 +1,160 @@
+dataset_info = dict(
+    dataset_name='panoptic_pose_3d',
+    paper_info=dict(
+        author='Joo, Hanbyul and Simon, Tomas and  Li, Xulong'
+        'and Liu, Hao and Tan, Lei and Gui, Lin and Banerjee, Sean'
+        'and Godisart, Timothy and Nabbe, Bart and Matthews, Iain'
+        'and Kanade, Takeo and Nobuhara, Shohei and Sheikh, Yaser',
+        title='Panoptic Studio: A Massively Multiview System '
+        'for Interaction Motion Capture',
+        container='IEEE Transactions on Pattern Analysis'
+        ' and Machine Intelligence',
+        year='2017',
+        homepage='http://domedb.perception.cs.cmu.edu',
+    ),
+    keypoint_info={
+        0:
+        dict(name='neck', id=0, color=[51, 153, 255], type='upper', swap=''),
+        1:
+        dict(name='nose', id=1, color=[51, 153, 255], type='upper', swap=''),
+        2:
+        dict(name='mid_hip', id=2, color=[0, 255, 0], type='lower', swap=''),
+        3:
+        dict(
+            name='left_shoulder',
+            id=3,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        4:
+        dict(
+            name='left_elbow',
+            id=4,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        5:
+        dict(
+            name='left_wrist',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        6:
+        dict(
+            name='left_hip',
+            id=6,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        7:
+        dict(
+            name='left_knee',
+            id=7,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        8:
+        dict(
+            name='left_ankle',
+            id=8,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        9:
+        dict(
+            name='right_shoulder',
+            id=9,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        10:
+        dict(
+            name='right_elbow',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        11:
+        dict(
+            name='right_wrist',
+            id=11,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        12:
+        dict(
+            name='right_hip',
+            id=12,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        13:
+        dict(
+            name='right_knee',
+            id=13,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        14:
+        dict(
+            name='right_ankle',
+            id=14,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle'),
+        15:
+        dict(
+            name='left_eye',
+            id=15,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_eye'),
+        16:
+        dict(
+            name='left_ear',
+            id=16,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_ear'),
+        17:
+        dict(
+            name='right_eye',
+            id=17,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_eye'),
+        18:
+        dict(
+            name='right_ear',
+            id=18,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_ear')
+    },
+    skeleton_info={
+        0: dict(link=('nose', 'neck'), id=0, color=[51, 153, 255]),
+        1: dict(link=('neck', 'left_shoulder'), id=1, color=[0, 255, 0]),
+        2: dict(link=('neck', 'right_shoulder'), id=2, color=[255, 128, 0]),
+        3: dict(link=('left_shoulder', 'left_elbow'), id=3, color=[0, 255, 0]),
+        4: dict(
+            link=('right_shoulder', 'right_elbow'), id=4, color=[255, 128, 0]),
+        5: dict(link=('left_elbow', 'left_wrist'), id=5, color=[0, 255, 0]),
+        6:
+        dict(link=('right_elbow', 'right_wrist'), id=6, color=[255, 128, 0]),
+        7: dict(link=('left_ankle', 'left_knee'), id=7, color=[0, 255, 0]),
+        8: dict(link=('left_knee', 'left_hip'), id=8, color=[0, 255, 0]),
+        9: dict(link=('right_ankle', 'right_knee'), id=9, color=[255, 128, 0]),
+        10: dict(link=('right_knee', 'right_hip'), id=10, color=[255, 128, 0]),
+        11: dict(link=('mid_hip', 'left_hip'), id=11, color=[0, 255, 0]),
+        12: dict(link=('mid_hip', 'right_hip'), id=12, color=[255, 128, 0]),
+        13: dict(link=('mid_hip', 'neck'), id=13, color=[51, 153, 255]),
+    },
+    joint_weights=[
+        1.0, 1.0, 1.0, 1.0, 1.2, 1.5, 1.0, 1.2, 1.5, 1.0, 1.2, 1.5, 1.0, 1.2,
+        1.5, 1.0, 1.0, 1.0, 1.0
+    ],
+    sigmas=[
+        0.026, 0.026, 0.107, 0.079, 0.072, 0.062, 0.107, 0.087, 0.089, 0.079,
+        0.072, 0.062, 0.107, 0.087, 0.089, 0.025, 0.035, 0.025, 0.035
+    ])

configs/_base_/datasets/panoptic_hand2d.py

@@ -0,0 +1,143 @@
+dataset_info = dict(
+    dataset_name='panoptic_hand2d',
+    paper_info=dict(
+        author='Simon, Tomas and Joo, Hanbyul and '
+        'Matthews, Iain and Sheikh, Yaser',
+        title='Hand keypoint detection in single images using '
+        'multiview bootstrapping',
+        container='Proceedings of the IEEE conference on '
+        'Computer Vision and Pattern Recognition',
+        year='2017',
+        homepage='http://domedb.perception.cs.cmu.edu/handdb.html',
+    ),
+    keypoint_info={
+        0:
+        dict(name='wrist', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='thumb1', id=1, color=[255, 128, 0], type='', swap=''),
+        2:
+        dict(name='thumb2', id=2, color=[255, 128, 0], type='', swap=''),
+        3:
+        dict(name='thumb3', id=3, color=[255, 128, 0], type='', swap=''),
+        4:
+        dict(name='thumb4', id=4, color=[255, 128, 0], type='', swap=''),
+        5:
+        dict(
+            name='forefinger1', id=5, color=[255, 153, 255], type='', swap=''),
+        6:
+        dict(
+            name='forefinger2', id=6, color=[255, 153, 255], type='', swap=''),
+        7:
+        dict(
+            name='forefinger3', id=7, color=[255, 153, 255], type='', swap=''),
+        8:
+        dict(
+            name='forefinger4', id=8, color=[255, 153, 255], type='', swap=''),
+        9:
+        dict(
+            name='middle_finger1',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        10:
+        dict(
+            name='middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        11:
+        dict(
+            name='middle_finger3',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        12:
+        dict(
+            name='middle_finger4',
+            id=12,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        13:
+        dict(
+            name='ring_finger1', id=13, color=[255, 51, 51], type='', swap=''),
+        14:
+        dict(
+            name='ring_finger2', id=14, color=[255, 51, 51], type='', swap=''),
+        15:
+        dict(
+            name='ring_finger3', id=15, color=[255, 51, 51], type='', swap=''),
+        16:
+        dict(
+            name='ring_finger4', id=16, color=[255, 51, 51], type='', swap=''),
+        17:
+        dict(name='pinky_finger1', id=17, color=[0, 255, 0], type='', swap=''),
+        18:
+        dict(name='pinky_finger2', id=18, color=[0, 255, 0], type='', swap=''),
+        19:
+        dict(name='pinky_finger3', id=19, color=[0, 255, 0], type='', swap=''),
+        20:
+        dict(name='pinky_finger4', id=20, color=[0, 255, 0], type='', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('wrist', 'thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('thumb1', 'thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('thumb2', 'thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('thumb3', 'thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('wrist', 'forefinger1'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('forefinger1', 'forefinger2'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('forefinger2', 'forefinger3'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('forefinger3', 'forefinger4'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('wrist', 'middle_finger1'), id=8, color=[102, 178, 255]),
+        9:
+        dict(
+            link=('middle_finger1', 'middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('middle_finger2', 'middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('middle_finger3', 'middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(link=('wrist', 'ring_finger1'), id=12, color=[255, 51, 51]),
+        13:
+        dict(
+            link=('ring_finger1', 'ring_finger2'), id=13, color=[255, 51, 51]),
+        14:
+        dict(
+            link=('ring_finger2', 'ring_finger3'), id=14, color=[255, 51, 51]),
+        15:
+        dict(
+            link=('ring_finger3', 'ring_finger4'), id=15, color=[255, 51, 51]),
+        16:
+        dict(link=('wrist', 'pinky_finger1'), id=16, color=[0, 255, 0]),
+        17:
+        dict(
+            link=('pinky_finger1', 'pinky_finger2'), id=17, color=[0, 255, 0]),
+        18:
+        dict(
+            link=('pinky_finger2', 'pinky_finger3'), id=18, color=[0, 255, 0]),
+        19:
+        dict(
+            link=('pinky_finger3', 'pinky_finger4'), id=19, color=[0, 255, 0])
+    },
+    joint_weights=[1.] * 21,
+    sigmas=[])

configs/_base_/datasets/posetrack18.py

@@ -0,0 +1,176 @@
+dataset_info = dict(
+    dataset_name='posetrack18',
+    paper_info=dict(
+        author='Andriluka, Mykhaylo and Iqbal, Umar and '
+        'Insafutdinov, Eldar and Pishchulin, Leonid and '
+        'Milan, Anton and Gall, Juergen and Schiele, Bernt',
+        title='Posetrack: A benchmark for human pose estimation and tracking',
+        container='Proceedings of the IEEE Conference on '
+        'Computer Vision and Pattern Recognition',
+        year='2018',
+        homepage='https://posetrack.net/users/download.php',
+    ),
+    keypoint_info={
+        0:
+        dict(name='nose', id=0, color=[51, 153, 255], type='upper', swap=''),
+        1:
+        dict(
+            name='head_bottom',
+            id=1,
+            color=[51, 153, 255],
+            type='upper',
+            swap=''),
+        2:
+        dict(
+            name='head_top', id=2, color=[51, 153, 255], type='upper',
+            swap=''),
+        3:
+        dict(
+            name='left_ear',
+            id=3,
+            color=[51, 153, 255],
+            type='upper',
+            swap='right_ear'),
+        4:
+        dict(
+            name='right_ear',
+            id=4,
+            color=[51, 153, 255],
+            type='upper',
+            swap='left_ear'),
+        5:
+        dict(
+            name='left_shoulder',
+            id=5,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_shoulder'),
+        6:
+        dict(
+            name='right_shoulder',
+            id=6,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_shoulder'),
+        7:
+        dict(
+            name='left_elbow',
+            id=7,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_elbow'),
+        8:
+        dict(
+            name='right_elbow',
+            id=8,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_elbow'),
+        9:
+        dict(
+            name='left_wrist',
+            id=9,
+            color=[0, 255, 0],
+            type='upper',
+            swap='right_wrist'),
+        10:
+        dict(
+            name='right_wrist',
+            id=10,
+            color=[255, 128, 0],
+            type='upper',
+            swap='left_wrist'),
+        11:
+        dict(
+            name='left_hip',
+            id=11,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_hip'),
+        12:
+        dict(
+            name='right_hip',
+            id=12,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_hip'),
+        13:
+        dict(
+            name='left_knee',
+            id=13,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_knee'),
+        14:
+        dict(
+            name='right_knee',
+            id=14,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_knee'),
+        15:
+        dict(
+            name='left_ankle',
+            id=15,
+            color=[0, 255, 0],
+            type='lower',
+            swap='right_ankle'),
+        16:
+        dict(
+            name='right_ankle',
+            id=16,
+            color=[255, 128, 0],
+            type='lower',
+            swap='left_ankle')
+    },
+    skeleton_info={
+        0:
+        dict(link=('left_ankle', 'left_knee'), id=0, color=[0, 255, 0]),
+        1:
+        dict(link=('left_knee', 'left_hip'), id=1, color=[0, 255, 0]),
+        2:
+        dict(link=('right_ankle', 'right_knee'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('right_knee', 'right_hip'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('left_hip', 'right_hip'), id=4, color=[51, 153, 255]),
+        5:
+        dict(link=('left_shoulder', 'left_hip'), id=5, color=[51, 153, 255]),
+        6:
+        dict(link=('right_shoulder', 'right_hip'), id=6, color=[51, 153, 255]),
+        7:
+        dict(
+            link=('left_shoulder', 'right_shoulder'),
+            id=7,
+            color=[51, 153, 255]),
+        8:
+        dict(link=('left_shoulder', 'left_elbow'), id=8, color=[0, 255, 0]),
+        9:
+        dict(
+            link=('right_shoulder', 'right_elbow'), id=9, color=[255, 128, 0]),
+        10:
+        dict(link=('left_elbow', 'left_wrist'), id=10, color=[0, 255, 0]),
+        11:
+        dict(link=('right_elbow', 'right_wrist'), id=11, color=[255, 128, 0]),
+        12:
+        dict(link=('nose', 'head_bottom'), id=12, color=[51, 153, 255]),
+        13:
+        dict(link=('nose', 'head_top'), id=13, color=[51, 153, 255]),
+        14:
+        dict(
+            link=('head_bottom', 'left_shoulder'), id=14, color=[51, 153,
+                                                                 255]),
+        15:
+        dict(
+            link=('head_bottom', 'right_shoulder'),
+            id=15,
+            color=[51, 153, 255])
+    },
+    joint_weights=[
+        1., 1., 1., 1., 1., 1., 1., 1.2, 1.2, 1.5, 1.5, 1., 1., 1.2, 1.2, 1.5,
+        1.5
+    ],
+    sigmas=[
+        0.026, 0.025, 0.025, 0.035, 0.035, 0.079, 0.079, 0.072, 0.072, 0.062,
+        0.062, 0.107, 0.107, 0.087, 0.087, 0.089, 0.089
+    ])

configs/_base_/datasets/rhd2d.py

@@ -0,0 +1,141 @@
+dataset_info = dict(
+    dataset_name='rhd2d',
+    paper_info=dict(
+        author='Christian Zimmermann and Thomas Brox',
+        title='Learning to Estimate 3D Hand Pose from Single RGB Images',
+        container='arXiv',
+        year='2017',
+        homepage='https://lmb.informatik.uni-freiburg.de/resources/'
+        'datasets/RenderedHandposeDataset.en.html',
+    ),
+    keypoint_info={
+        0:
+        dict(name='wrist', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='thumb1', id=1, color=[255, 128, 0], type='', swap=''),
+        2:
+        dict(name='thumb2', id=2, color=[255, 128, 0], type='', swap=''),
+        3:
+        dict(name='thumb3', id=3, color=[255, 128, 0], type='', swap=''),
+        4:
+        dict(name='thumb4', id=4, color=[255, 128, 0], type='', swap=''),
+        5:
+        dict(
+            name='forefinger1', id=5, color=[255, 153, 255], type='', swap=''),
+        6:
+        dict(
+            name='forefinger2', id=6, color=[255, 153, 255], type='', swap=''),
+        7:
+        dict(
+            name='forefinger3', id=7, color=[255, 153, 255], type='', swap=''),
+        8:
+        dict(
+            name='forefinger4', id=8, color=[255, 153, 255], type='', swap=''),
+        9:
+        dict(
+            name='middle_finger1',
+            id=9,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        10:
+        dict(
+            name='middle_finger2',
+            id=10,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        11:
+        dict(
+            name='middle_finger3',
+            id=11,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        12:
+        dict(
+            name='middle_finger4',
+            id=12,
+            color=[102, 178, 255],
+            type='',
+            swap=''),
+        13:
+        dict(
+            name='ring_finger1', id=13, color=[255, 51, 51], type='', swap=''),
+        14:
+        dict(
+            name='ring_finger2', id=14, color=[255, 51, 51], type='', swap=''),
+        15:
+        dict(
+            name='ring_finger3', id=15, color=[255, 51, 51], type='', swap=''),
+        16:
+        dict(
+            name='ring_finger4', id=16, color=[255, 51, 51], type='', swap=''),
+        17:
+        dict(name='pinky_finger1', id=17, color=[0, 255, 0], type='', swap=''),
+        18:
+        dict(name='pinky_finger2', id=18, color=[0, 255, 0], type='', swap=''),
+        19:
+        dict(name='pinky_finger3', id=19, color=[0, 255, 0], type='', swap=''),
+        20:
+        dict(name='pinky_finger4', id=20, color=[0, 255, 0], type='', swap='')
+    },
+    skeleton_info={
+        0:
+        dict(link=('wrist', 'thumb1'), id=0, color=[255, 128, 0]),
+        1:
+        dict(link=('thumb1', 'thumb2'), id=1, color=[255, 128, 0]),
+        2:
+        dict(link=('thumb2', 'thumb3'), id=2, color=[255, 128, 0]),
+        3:
+        dict(link=('thumb3', 'thumb4'), id=3, color=[255, 128, 0]),
+        4:
+        dict(link=('wrist', 'forefinger1'), id=4, color=[255, 153, 255]),
+        5:
+        dict(link=('forefinger1', 'forefinger2'), id=5, color=[255, 153, 255]),
+        6:
+        dict(link=('forefinger2', 'forefinger3'), id=6, color=[255, 153, 255]),
+        7:
+        dict(link=('forefinger3', 'forefinger4'), id=7, color=[255, 153, 255]),
+        8:
+        dict(link=('wrist', 'middle_finger1'), id=8, color=[102, 178, 255]),
+        9:
+        dict(
+            link=('middle_finger1', 'middle_finger2'),
+            id=9,
+            color=[102, 178, 255]),
+        10:
+        dict(
+            link=('middle_finger2', 'middle_finger3'),
+            id=10,
+            color=[102, 178, 255]),
+        11:
+        dict(
+            link=('middle_finger3', 'middle_finger4'),
+            id=11,
+            color=[102, 178, 255]),
+        12:
+        dict(link=('wrist', 'ring_finger1'), id=12, color=[255, 51, 51]),
+        13:
+        dict(
+            link=('ring_finger1', 'ring_finger2'), id=13, color=[255, 51, 51]),
+        14:
+        dict(
+            link=('ring_finger2', 'ring_finger3'), id=14, color=[255, 51, 51]),
+        15:
+        dict(
+            link=('ring_finger3', 'ring_finger4'), id=15, color=[255, 51, 51]),
+        16:
+        dict(link=('wrist', 'pinky_finger1'), id=16, color=[0, 255, 0]),
+        17:
+        dict(
+            link=('pinky_finger1', 'pinky_finger2'), id=17, color=[0, 255, 0]),
+        18:
+        dict(
+            link=('pinky_finger2', 'pinky_finger3'), id=18, color=[0, 255, 0]),
+        19:
+        dict(
+            link=('pinky_finger3', 'pinky_finger4'), id=19, color=[0, 255, 0])
+    },
+    joint_weights=[1.] * 21,
+    sigmas=[])

configs/_base_/datasets/wflw.py

@@ -0,0 +1,582 @@
+dataset_info = dict(
+    dataset_name='wflw',
+    paper_info=dict(
+        author='Wu, Wayne and Qian, Chen and Yang, Shuo and Wang, '
+        'Quan and Cai, Yici and Zhou, Qiang',
+        title='Look at boundary: A boundary-aware face alignment algorithm',
+        container='Proceedings of the IEEE conference on computer '
+        'vision and pattern recognition',
+        year='2018',
+        homepage='https://wywu.github.io/projects/LAB/WFLW.html',
+    ),
+    keypoint_info={
+        0:
+        dict(
+            name='kpt-0', id=0, color=[255, 255, 255], type='', swap='kpt-32'),
+        1:
+        dict(
+            name='kpt-1', id=1, color=[255, 255, 255], type='', swap='kpt-31'),
+        2:
+        dict(
+            name='kpt-2', id=2, color=[255, 255, 255], type='', swap='kpt-30'),
+        3:
+        dict(
+            name='kpt-3', id=3, color=[255, 255, 255], type='', swap='kpt-29'),
+        4:
+        dict(
+            name='kpt-4', id=4, color=[255, 255, 255], type='', swap='kpt-28'),
+        5:
+        dict(
+            name='kpt-5', id=5, color=[255, 255, 255], type='', swap='kpt-27'),
+        6:
+        dict(
+            name='kpt-6', id=6, color=[255, 255, 255], type='', swap='kpt-26'),
+        7:
+        dict(
+            name='kpt-7', id=7, color=[255, 255, 255], type='', swap='kpt-25'),
+        8:
+        dict(
+            name='kpt-8', id=8, color=[255, 255, 255], type='', swap='kpt-24'),
+        9:
+        dict(
+            name='kpt-9', id=9, color=[255, 255, 255], type='', swap='kpt-23'),
+        10:
+        dict(
+            name='kpt-10',
+            id=10,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-22'),
+        11:
+        dict(
+            name='kpt-11',
+            id=11,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-21'),
+        12:
+        dict(
+            name='kpt-12',
+            id=12,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-20'),
+        13:
+        dict(
+            name='kpt-13',
+            id=13,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-19'),
+        14:
+        dict(
+            name='kpt-14',
+            id=14,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-18'),
+        15:
+        dict(
+            name='kpt-15',
+            id=15,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-17'),
+        16:
+        dict(name='kpt-16', id=16, color=[255, 255, 255], type='', swap=''),
+        17:
+        dict(
+            name='kpt-17',
+            id=17,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-15'),
+        18:
+        dict(
+            name='kpt-18',
+            id=18,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-14'),
+        19:
+        dict(
+            name='kpt-19',
+            id=19,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-13'),
+        20:
+        dict(
+            name='kpt-20',
+            id=20,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-12'),
+        21:
+        dict(
+            name='kpt-21',
+            id=21,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-11'),
+        22:
+        dict(
+            name='kpt-22',
+            id=22,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-10'),
+        23:
+        dict(
+            name='kpt-23', id=23, color=[255, 255, 255], type='',
+            swap='kpt-9'),
+        24:
+        dict(
+            name='kpt-24', id=24, color=[255, 255, 255], type='',
+            swap='kpt-8'),
+        25:
+        dict(
+            name='kpt-25', id=25, color=[255, 255, 255], type='',
+            swap='kpt-7'),
+        26:
+        dict(
+            name='kpt-26', id=26, color=[255, 255, 255], type='',
+            swap='kpt-6'),
+        27:
+        dict(
+            name='kpt-27', id=27, color=[255, 255, 255], type='',
+            swap='kpt-5'),
+        28:
+        dict(
+            name='kpt-28', id=28, color=[255, 255, 255], type='',
+            swap='kpt-4'),
+        29:
+        dict(
+            name='kpt-29', id=29, color=[255, 255, 255], type='',
+            swap='kpt-3'),
+        30:
+        dict(
+            name='kpt-30', id=30, color=[255, 255, 255], type='',
+            swap='kpt-2'),
+        31:
+        dict(
+            name='kpt-31', id=31, color=[255, 255, 255], type='',
+            swap='kpt-1'),
+        32:
+        dict(
+            name='kpt-32', id=32, color=[255, 255, 255], type='',
+            swap='kpt-0'),
+        33:
+        dict(
+            name='kpt-33',
+            id=33,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-46'),
+        34:
+        dict(
+            name='kpt-34',
+            id=34,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-45'),
+        35:
+        dict(
+            name='kpt-35',
+            id=35,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-44'),
+        36:
+        dict(
+            name='kpt-36',
+            id=36,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-43'),
+        37:
+        dict(
+            name='kpt-37',
+            id=37,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-42'),
+        38:
+        dict(
+            name='kpt-38',
+            id=38,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-50'),
+        39:
+        dict(
+            name='kpt-39',
+            id=39,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-49'),
+        40:
+        dict(
+            name='kpt-40',
+            id=40,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-48'),
+        41:
+        dict(
+            name='kpt-41',
+            id=41,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-47'),
+        42:
+        dict(
+            name='kpt-42',
+            id=42,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-37'),
+        43:
+        dict(
+            name='kpt-43',
+            id=43,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-36'),
+        44:
+        dict(
+            name='kpt-44',
+            id=44,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-35'),
+        45:
+        dict(
+            name='kpt-45',
+            id=45,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-34'),
+        46:
+        dict(
+            name='kpt-46',
+            id=46,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-33'),
+        47:
+        dict(
+            name='kpt-47',
+            id=47,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-41'),
+        48:
+        dict(
+            name='kpt-48',
+            id=48,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-40'),
+        49:
+        dict(
+            name='kpt-49',
+            id=49,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-39'),
+        50:
+        dict(
+            name='kpt-50',
+            id=50,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-38'),
+        51:
+        dict(name='kpt-51', id=51, color=[255, 255, 255], type='', swap=''),
+        52:
+        dict(name='kpt-52', id=52, color=[255, 255, 255], type='', swap=''),
+        53:
+        dict(name='kpt-53', id=53, color=[255, 255, 255], type='', swap=''),
+        54:
+        dict(name='kpt-54', id=54, color=[255, 255, 255], type='', swap=''),
+        55:
+        dict(
+            name='kpt-55',
+            id=55,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-59'),
+        56:
+        dict(
+            name='kpt-56',
+            id=56,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-58'),
+        57:
+        dict(name='kpt-57', id=57, color=[255, 255, 255], type='', swap=''),
+        58:
+        dict(
+            name='kpt-58',
+            id=58,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-56'),
+        59:
+        dict(
+            name='kpt-59',
+            id=59,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-55'),
+        60:
+        dict(
+            name='kpt-60',
+            id=60,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-72'),
+        61:
+        dict(
+            name='kpt-61',
+            id=61,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-71'),
+        62:
+        dict(
+            name='kpt-62',
+            id=62,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-70'),
+        63:
+        dict(
+            name='kpt-63',
+            id=63,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-69'),
+        64:
+        dict(
+            name='kpt-64',
+            id=64,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-68'),
+        65:
+        dict(
+            name='kpt-65',
+            id=65,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-75'),
+        66:
+        dict(
+            name='kpt-66',
+            id=66,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-74'),
+        67:
+        dict(
+            name='kpt-67',
+            id=67,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-73'),
+        68:
+        dict(
+            name='kpt-68',
+            id=68,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-64'),
+        69:
+        dict(
+            name='kpt-69',
+            id=69,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-63'),
+        70:
+        dict(
+            name='kpt-70',
+            id=70,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-62'),
+        71:
+        dict(
+            name='kpt-71',
+            id=71,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-61'),
+        72:
+        dict(
+            name='kpt-72',
+            id=72,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-60'),
+        73:
+        dict(
+            name='kpt-73',
+            id=73,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-67'),
+        74:
+        dict(
+            name='kpt-74',
+            id=74,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-66'),
+        75:
+        dict(
+            name='kpt-75',
+            id=75,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-65'),
+        76:
+        dict(
+            name='kpt-76',
+            id=76,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-82'),
+        77:
+        dict(
+            name='kpt-77',
+            id=77,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-81'),
+        78:
+        dict(
+            name='kpt-78',
+            id=78,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-80'),
+        79:
+        dict(name='kpt-79', id=79, color=[255, 255, 255], type='', swap=''),
+        80:
+        dict(
+            name='kpt-80',
+            id=80,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-78'),
+        81:
+        dict(
+            name='kpt-81',
+            id=81,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-77'),
+        82:
+        dict(
+            name='kpt-82',
+            id=82,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-76'),
+        83:
+        dict(
+            name='kpt-83',
+            id=83,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-87'),
+        84:
+        dict(
+            name='kpt-84',
+            id=84,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-86'),
+        85:
+        dict(name='kpt-85', id=85, color=[255, 255, 255], type='', swap=''),
+        86:
+        dict(
+            name='kpt-86',
+            id=86,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-84'),
+        87:
+        dict(
+            name='kpt-87',
+            id=87,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-83'),
+        88:
+        dict(
+            name='kpt-88',
+            id=88,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-92'),
+        89:
+        dict(
+            name='kpt-89',
+            id=89,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-91'),
+        90:
+        dict(name='kpt-90', id=90, color=[255, 255, 255], type='', swap=''),
+        91:
+        dict(
+            name='kpt-91',
+            id=91,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-89'),
+        92:
+        dict(
+            name='kpt-92',
+            id=92,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-88'),
+        93:
+        dict(
+            name='kpt-93',
+            id=93,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-95'),
+        94:
+        dict(name='kpt-94', id=94, color=[255, 255, 255], type='', swap=''),
+        95:
+        dict(
+            name='kpt-95',
+            id=95,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-93'),
+        96:
+        dict(
+            name='kpt-96',
+            id=96,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-97'),
+        97:
+        dict(
+            name='kpt-97',
+            id=97,
+            color=[255, 255, 255],
+            type='',
+            swap='kpt-96')
+    },
+    skeleton_info={},
+    joint_weights=[1.] * 98,
+    sigmas=[])

configs/_base_/datasets/zebra.py

@@ -0,0 +1,64 @@
+dataset_info = dict(
+    dataset_name='zebra',
+    paper_info=dict(
+        author='Graving, Jacob M and Chae, Daniel and Naik, Hemal and '
+        'Li, Liang and Koger, Benjamin and Costelloe, Blair R and '
+        'Couzin, Iain D',
+        title='DeepPoseKit, a software toolkit for fast and robust '
+        'animal pose estimation using deep learning',
+        container='Elife',
+        year='2019',
+        homepage='https://github.com/jgraving/DeepPoseKit-Data',
+    ),
+    keypoint_info={
+        0:
+        dict(name='snout', id=0, color=[255, 255, 255], type='', swap=''),
+        1:
+        dict(name='head', id=1, color=[255, 255, 255], type='', swap=''),
+        2:
+        dict(name='neck', id=2, color=[255, 255, 255], type='', swap=''),
+        3:
+        dict(
+            name='forelegL1',
+            id=3,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegR1'),
+        4:
+        dict(
+            name='forelegR1',
+            id=4,
+            color=[255, 255, 255],
+            type='',
+            swap='forelegL1'),
+        5:
+        dict(
+            name='hindlegL1',
+            id=5,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegR1'),
+        6:
+        dict(
+            name='hindlegR1',
+            id=6,
+            color=[255, 255, 255],
+            type='',
+            swap='hindlegL1'),
+        7:
+        dict(name='tailbase', id=7, color=[255, 255, 255], type='', swap=''),
+        8:
+        dict(name='tailtip', id=8, color=[255, 255, 255], type='', swap='')
+    },
+    skeleton_info={
+        0: dict(link=('head', 'snout'), id=0, color=[255, 255, 255]),
+        1: dict(link=('neck', 'head'), id=1, color=[255, 255, 255]),
+        2: dict(link=('forelegL1', 'neck'), id=2, color=[255, 255, 255]),
+        3: dict(link=('forelegR1', 'neck'), id=3, color=[255, 255, 255]),
+        4: dict(link=('hindlegL1', 'tailbase'), id=4, color=[255, 255, 255]),
+        5: dict(link=('hindlegR1', 'tailbase'), id=5, color=[255, 255, 255]),
+        6: dict(link=('tailbase', 'neck'), id=6, color=[255, 255, 255]),
+        7: dict(link=('tailtip', 'tailbase'), id=7, color=[255, 255, 255])
+    },
+    joint_weights=[1.] * 9,
+    sigmas=[])

configs/_base_/default_runtime.py

@@ -0,0 +1,19 @@
+checkpoint_config = dict(interval=10)
+
+log_config = dict(
+    interval=50,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+log_level = 'INFO'
+load_from = None
+resume_from = None
+dist_params = dict(backend='nccl')
+workflow = [('train', 1)]
+
+# disable opencv multithreading to avoid system being overloaded
+opencv_num_threads = 0
+# set multi-process start method as `fork` to speed up the training
+mp_start_method = 'fork'

configs/animal/2d_kpt_sview_rgb_img/README.md

@@ -0,0 +1,18 @@
+# 2D Animal Keypoint Detection
+
+2D animal keypoint detection (animal pose estimation) aims to detect the key-point of different species, including rats,
+dogs, macaques, and cheetah. It provides detailed behavioral analysis for neuroscience, medical and ecology applications.
+
+## Data preparation
+
+Please follow [DATA Preparation](/docs/en/tasks/2d_animal_keypoint.md) to prepare data.
+
+## Demo
+
+Please follow [DEMO](/demo/docs/2d_animal_demo.md) to generate fancy demos.
+
+<img src="https://user-images.githubusercontent.com/11788150/114201893-4446ec00-9989-11eb-808b-5718c47c7b23.gif" height="140px" alt><br>
+
+<img src="https://user-images.githubusercontent.com/11788150/114205282-b5d46980-998c-11eb-9d6b-85ba47f81252.gif" height="140px" alt><br>
+
+<img src="https://user-images.githubusercontent.com/11788150/114023530-944c8280-98a5-11eb-86b0-5f6d3e232af0.gif" height="140px" alt><br>

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/README.md

@@ -0,0 +1,7 @@
+# Top-down heatmap-based pose estimation
+
+Top-down methods divide the task into two stages: object detection and pose estimation.
+
+They perform object detection first, followed by single-object pose estimation given object bounding boxes.
+Instead of estimating keypoint coordinates directly, the pose estimator will produce heatmaps which represent the
+likelihood of being a keypoint.

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_animalpose.md

@@ -0,0 +1,40 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html">HRNet (CVPR'2019)</a></summary>
+
+```bibtex
+@inproceedings{sun2019deep,
+  title={Deep high-resolution representation learning for human pose estimation},
+  author={Sun, Ke and Xiao, Bin and Liu, Dong and Wang, Jingdong},
+  booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
+  pages={5693--5703},
+  year={2019}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ICCV_2019/html/Cao_Cross-Domain_Adaptation_for_Animal_Pose_Estimation_ICCV_2019_paper.html">Animal-Pose (ICCV'2019)</a></summary>
+
+```bibtex
+@InProceedings{Cao_2019_ICCV,
+    author = {Cao, Jinkun and Tang, Hongyang and Fang, Hao-Shu and Shen, Xiaoyong and Lu, Cewu and Tai, Yu-Wing},
+    title = {Cross-Domain Adaptation for Animal Pose Estimation},
+    booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
+    month = {October},
+    year = {2019}
+}
+```
+
+</details>
+
+Results on AnimalPose validation set (1117 instances)
+
+| Arch  | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
+| :-------------- | :-----------: | :------: | :------: | :------: | :------: | :------: |:------: |:------: |
+| [pose_hrnet_w32](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_w32_animalpose_256x256.py)  | 256x256 | 0.736 | 0.959 | 0.832 | 0.775 | 0.966 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_animalpose_256x256-1aa7f075_20210426.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_animalpose_256x256_20210426.log.json) |
+| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_w48_animalpose_256x256.py)  | 256x256 | 0.737 | 0.959 | 0.823 | 0.778 | 0.962 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_animalpose_256x256-34644726_20210426.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_animalpose_256x256_20210426.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_animalpose.yml

@@ -0,0 +1,40 @@
+Collections:
+- Name: HRNet
+  Paper:
+    Title: Deep high-resolution representation learning for human pose estimation
+    URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/backbones/hrnet.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_w32_animalpose_256x256.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: &id001
+    - HRNet
+    Training Data: Animal-Pose
+  Name: topdown_heatmap_hrnet_w32_animalpose_256x256
+  Results:
+  - Dataset: Animal-Pose
+    Metrics:
+      AP: 0.736
+      AP@0.5: 0.959
+      AP@0.75: 0.832
+      AR: 0.775
+      AR@0.5: 0.966
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_animalpose_256x256-1aa7f075_20210426.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_w48_animalpose_256x256.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: Animal-Pose
+  Name: topdown_heatmap_hrnet_w48_animalpose_256x256
+  Results:
+  - Dataset: Animal-Pose
+    Metrics:
+      AP: 0.737
+      AP@0.5: 0.959
+      AP@0.75: 0.823
+      AR: 0.778
+      AR@0.5: 0.962
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_animalpose_256x256-34644726_20210426.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_w32_animalpose_256x256.py

@@ -0,0 +1,172 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/animalpose.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=20,
+    dataset_joints=20,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/animalpose'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_train.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/hrnet_w48_animalpose_256x256.py

@@ -0,0 +1,172 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/animalpose.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=20,
+    dataset_joints=20,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w48-8ef0771d.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(48, 96)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(48, 96, 192)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(48, 96, 192, 384))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/animalpose'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_train.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res101_animalpose_256x256.py

@@ -0,0 +1,141 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/animalpose.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=20,
+    dataset_joints=20,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/animalpose'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_train.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res152_animalpose_256x256.py

@@ -0,0 +1,141 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/animalpose.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=20,
+    dataset_joints=20,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/animalpose'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_train.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res50_animalpose_256x256.py

@@ -0,0 +1,141 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/animalpose.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=20,
+    dataset_joints=20,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/animalpose'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_train.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalPoseDataset',
+        ann_file=f'{data_root}/annotations/animalpose_val.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/resnet_animalpose.md

@@ -0,0 +1,41 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ICCV_2019/html/Cao_Cross-Domain_Adaptation_for_Animal_Pose_Estimation_ICCV_2019_paper.html">Animal-Pose (ICCV'2019)</a></summary>
+
+```bibtex
+@InProceedings{Cao_2019_ICCV,
+    author = {Cao, Jinkun and Tang, Hongyang and Fang, Hao-Shu and Shen, Xiaoyong and Lu, Cewu and Tai, Yu-Wing},
+    title = {Cross-Domain Adaptation for Animal Pose Estimation},
+    booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
+    month = {October},
+    year = {2019}
+}
+```
+
+</details>
+
+Results on AnimalPose validation set (1117 instances)
+
+| Arch  | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
+| :-------------- | :-----------: | :------: | :------: | :------: | :------: | :------: |:------: |:------: |
+| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res50_animalpose_256x256.py)  | 256x256 | 0.688 | 0.945 | 0.772 | 0.733 | 0.952 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_animalpose_256x256-e1f30bff_20210426.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_animalpose_256x256_20210426.log.json) |
+| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res101_animalpose_256x256.py) | 256x256 | 0.696 | 0.948 | 0.785 | 0.737 | 0.954 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_animalpose_256x256-85563f4a_20210426.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_animalpose_256x256_20210426.log.json) |
+| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res152_animalpose_256x256.py) | 256x256 | 0.709 | 0.948 | 0.797 | 0.749 | 0.951 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_animalpose_256x256-a0a7506c_20210426.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_animalpose_256x256_20210426.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/resnet_animalpose.yml

@@ -0,0 +1,56 @@
+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res50_animalpose_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: Animal-Pose
+  Name: topdown_heatmap_res50_animalpose_256x256
+  Results:
+  - Dataset: Animal-Pose
+    Metrics:
+      AP: 0.688
+      AP@0.5: 0.945
+      AP@0.75: 0.772
+      AR: 0.733
+      AR@0.5: 0.952
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_animalpose_256x256-e1f30bff_20210426.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res101_animalpose_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Animal-Pose
+  Name: topdown_heatmap_res101_animalpose_256x256
+  Results:
+  - Dataset: Animal-Pose
+    Metrics:
+      AP: 0.696
+      AP@0.5: 0.948
+      AP@0.75: 0.785
+      AR: 0.737
+      AR@0.5: 0.954
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_animalpose_256x256-85563f4a_20210426.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/animalpose/res152_animalpose_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Animal-Pose
+  Name: topdown_heatmap_res152_animalpose_256x256
+  Results:
+  - Dataset: Animal-Pose
+    Metrics:
+      AP: 0.709
+      AP@0.5: 0.948
+      AP@0.75: 0.797
+      AR: 0.749
+      AR@0.5: 0.951
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_animalpose_256x256-a0a7506c_20210426.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_ap10k.md

@@ -0,0 +1,41 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html">HRNet (CVPR'2019)</a></summary>
+
+```bibtex
+@inproceedings{sun2019deep,
+  title={Deep high-resolution representation learning for human pose estimation},
+  author={Sun, Ke and Xiao, Bin and Liu, Dong and Wang, Jingdong},
+  booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
+  pages={5693--5703},
+  year={2019}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://arxiv.org/abs/2108.12617">AP-10K (NeurIPS'2021)</a></summary>
+
+```bibtex
+@misc{yu2021ap10k,
+      title={AP-10K: A Benchmark for Animal Pose Estimation in the Wild},
+      author={Hang Yu and Yufei Xu and Jing Zhang and Wei Zhao and Ziyu Guan and Dacheng Tao},
+      year={2021},
+      eprint={2108.12617},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV}
+}
+```
+
+</details>
+
+Results on AP-10K validation set
+
+| Arch  | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AP<sup>M</sup> | AP<sup>L</sup> | ckpt | log |
+| :-------------- | :-----------: | :------: | :------: | :------: | :------: | :------: |:------: |:------: |
+| [pose_hrnet_w32](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_w32_ap10k_256x256.py)  | 256x256 | 0.738 | 0.958 | 0.808 | 0.592 | 0.743 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_ap10k_256x256-18aac840_20211029.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_ap10k_256x256-18aac840_20211029.log.json) |
+| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_w48_ap10k_256x256.py)  | 256x256 | 0.744 | 0.959 | 0.807 | 0.589 | 0.748 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_ap10k_256x256-d95ab412_20211029.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_ap10k_256x256-d95ab412_20211029.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_ap10k.yml

@@ -0,0 +1,40 @@
+Collections:
+- Name: HRNet
+  Paper:
+    Title: Deep high-resolution representation learning for human pose estimation
+    URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/backbones/hrnet.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_w32_ap10k_256x256.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: &id001
+    - HRNet
+    Training Data: AP-10K
+  Name: topdown_heatmap_hrnet_w32_ap10k_256x256
+  Results:
+  - Dataset: AP-10K
+    Metrics:
+      AP: 0.738
+      AP@0.5: 0.958
+      AP@0.75: 0.808
+      APL: 0.743
+      APM: 0.592
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_ap10k_256x256-18aac840_20211029.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_w48_ap10k_256x256.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: AP-10K
+  Name: topdown_heatmap_hrnet_w48_ap10k_256x256
+  Results:
+  - Dataset: AP-10K
+    Metrics:
+      AP: 0.744
+      AP@0.5: 0.959
+      AP@0.75: 0.807
+      APL: 0.748
+      APM: 0.589
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_ap10k_256x256-d95ab412_20211029.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_w32_ap10k_256x256.py

@@ -0,0 +1,172 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/ap10k.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=17,
+    dataset_joints=17,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/ap10k'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=4,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-train-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-val-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-test-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/hrnet_w48_ap10k_256x256.py

@@ -0,0 +1,172 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/ap10k.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=17,
+    dataset_joints=17,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w48-8ef0771d.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(48, 96)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(48, 96, 192)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(48, 96, 192, 384))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/ap10k'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=4,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-train-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-val-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-test-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/res101_ap10k_256x256.py

@@ -0,0 +1,141 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/ap10k.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=17,
+    dataset_joints=17,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/ap10k'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=4,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-train-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-val-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-test-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/res50_ap10k_256x256.py

@@ -0,0 +1,141 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/ap10k.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=17,
+    dataset_joints=17,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/ap10k'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=4,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-train-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-val-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalAP10KDataset',
+        ann_file=f'{data_root}/annotations/ap10k-test-split1.json',
+        img_prefix=f'{data_root}/data/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/resnet_ap10k.md

@@ -0,0 +1,41 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://arxiv.org/abs/2108.12617">AP-10K (NeurIPS'2021)</a></summary>
+
+```bibtex
+@misc{yu2021ap10k,
+      title={AP-10K: A Benchmark for Animal Pose Estimation in the Wild},
+      author={Hang Yu and Yufei Xu and Jing Zhang and Wei Zhao and Ziyu Guan and Dacheng Tao},
+      year={2021},
+      eprint={2108.12617},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV}
+}
+```
+
+</details>
+
+Results on AP-10K validation set
+
+| Arch  | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AP<sup>M</sup> | AP<sup>L</sup> | ckpt | log |
+| :-------------- | :-----------: | :------: | :------: | :------: | :------: | :------: |:------: |:------: |
+| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/res50_ap10k_256x256.py)  | 256x256 | 0.699 | 0.940 | 0.760 | 0.570 | 0.703 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_ap10k_256x256-35760eb8_20211029.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_ap10k_256x256-35760eb8_20211029.log.json) |
+| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/res101_ap10k_256x256.py) | 256x256 | 0.698 | 0.943 | 0.754 | 0.543 | 0.702 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_ap10k_256x256-9edfafb9_20211029.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_ap10k_256x256-9edfafb9_20211029.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/resnet_ap10k.yml

@@ -0,0 +1,40 @@
+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/res50_ap10k_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: AP-10K
+  Name: topdown_heatmap_res50_ap10k_256x256
+  Results:
+  - Dataset: AP-10K
+    Metrics:
+      AP: 0.699
+      AP@0.5: 0.94
+      AP@0.75: 0.76
+      APL: 0.703
+      APM: 0.57
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_ap10k_256x256-35760eb8_20211029.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/ap10k/res101_ap10k_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: AP-10K
+  Name: topdown_heatmap_res101_ap10k_256x256
+  Results:
+  - Dataset: AP-10K
+    Metrics:
+      AP: 0.698
+      AP@0.5: 0.943
+      AP@0.75: 0.754
+      APL: 0.702
+      APM: 0.543
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_ap10k_256x256-9edfafb9_20211029.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_atrw.md

@@ -0,0 +1,40 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html">HRNet (CVPR'2019)</a></summary>
+
+```bibtex
+@inproceedings{sun2019deep,
+  title={Deep high-resolution representation learning for human pose estimation},
+  author={Sun, Ke and Xiao, Bin and Liu, Dong and Wang, Jingdong},
+  booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
+  pages={5693--5703},
+  year={2019}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://arxiv.org/abs/1906.05586">ATRW (ACM MM'2020)</a></summary>
+
+```bibtex
+@inproceedings{li2020atrw,
+  title={ATRW: A Benchmark for Amur Tiger Re-identification in the Wild},
+  author={Li, Shuyuan and Li, Jianguo and Tang, Hanlin and Qian, Rui and Lin, Weiyao},
+  booktitle={Proceedings of the 28th ACM International Conference on Multimedia},
+  pages={2590--2598},
+  year={2020}
+}
+```
+
+</details>
+
+Results on ATRW validation set
+
+| Arch  | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
+| :-------------- | :-----------: | :------: | :------: | :------: | :------: | :------: |:------: |:------: |
+| [pose_hrnet_w32](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_w32_atrw_256x256.py)  | 256x256 | 0.912 | 0.973 | 0.959 | 0.938 | 0.985 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_atrw_256x256-f027f09a_20210414.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_atrw_256x256_20210414.log.json) |
+| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_w48_atrw_256x256.py)  | 256x256 | 0.911 | 0.972 | 0.946 | 0.937 | 0.985 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_atrw_256x256-ac088892_20210414.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_atrw_256x256_20210414.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_atrw.yml

@@ -0,0 +1,40 @@
+Collections:
+- Name: HRNet
+  Paper:
+    Title: Deep high-resolution representation learning for human pose estimation
+    URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/backbones/hrnet.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_w32_atrw_256x256.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: &id001
+    - HRNet
+    Training Data: ATRW
+  Name: topdown_heatmap_hrnet_w32_atrw_256x256
+  Results:
+  - Dataset: ATRW
+    Metrics:
+      AP: 0.912
+      AP@0.5: 0.973
+      AP@0.75: 0.959
+      AR: 0.938
+      AR@0.5: 0.985
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_atrw_256x256-f027f09a_20210414.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_w48_atrw_256x256.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: ATRW
+  Name: topdown_heatmap_hrnet_w48_atrw_256x256
+  Results:
+  - Dataset: ATRW
+    Metrics:
+      AP: 0.911
+      AP@0.5: 0.972
+      AP@0.75: 0.946
+      AR: 0.937
+      AR@0.5: 0.985
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_atrw_256x256-ac088892_20210414.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_w32_atrw_256x256.py

@@ -0,0 +1,170 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/atrw.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=15,
+    dataset_joints=15,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14],
+    ],
+    inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/atrw'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_train.json',
+        img_prefix=f'{data_root}/images/train/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/hrnet_w48_atrw_256x256.py

@@ -0,0 +1,170 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/atrw.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=15,
+    dataset_joints=15,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14],
+    ],
+    inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w48-8ef0771d.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(48, 96)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(48, 96, 192)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(48, 96, 192, 384))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/atrw'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_train.json',
+        img_prefix=f'{data_root}/images/train/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res101_atrw_256x256.py

@@ -0,0 +1,139 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/atrw.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=15,
+    dataset_joints=15,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14],
+    ],
+    inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/atrw'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_train.json',
+        img_prefix=f'{data_root}/images/train/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res152_atrw_256x256.py

@@ -0,0 +1,139 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/atrw.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=15,
+    dataset_joints=15,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14],
+    ],
+    inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/atrw'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_train.json',
+        img_prefix=f'{data_root}/images/train/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res50_atrw_256x256.py

@@ -0,0 +1,139 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/atrw.py'
+]
+evaluation = dict(interval=10, metric='mAP', save_best='AP')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=15,
+    dataset_joints=15,
+    dataset_channel=[
+        [0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14],
+    ],
+    inference_channel=[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'],
+    soft_nms=False,
+    nms_thr=1.0,
+    oks_thr=0.9,
+    vis_thr=0.2,
+    use_gt_bbox=True,
+    det_bbox_thr=0.0,
+    bbox_file='',
+)
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownHalfBodyTransform',
+        num_joints_half_body=8,
+        prob_half_body=0.3),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/atrw'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_train.json',
+        img_prefix=f'{data_root}/images/train/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalATRWDataset',
+        ann_file=f'{data_root}/annotations/keypoint_val.json',
+        img_prefix=f'{data_root}/images/val/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/resnet_atrw.md

@@ -0,0 +1,41 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://arxiv.org/abs/1906.05586">ATRW (ACM MM'2020)</a></summary>
+
+```bibtex
+@inproceedings{li2020atrw,
+  title={ATRW: A Benchmark for Amur Tiger Re-identification in the Wild},
+  author={Li, Shuyuan and Li, Jianguo and Tang, Hanlin and Qian, Rui and Lin, Weiyao},
+  booktitle={Proceedings of the 28th ACM International Conference on Multimedia},
+  pages={2590--2598},
+  year={2020}
+}
+```
+
+</details>
+
+Results on ATRW validation set
+
+| Arch  | Input Size | AP | AP<sup>50</sup> | AP<sup>75</sup> | AR | AR<sup>50</sup> | ckpt | log |
+| :-------------- | :-----------: | :------: | :------: | :------: | :------: | :------: |:------: |:------: |
+| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res50_atrw_256x256.py)  | 256x256 | 0.900 | 0.973 | 0.932 | 0.929 | 0.985 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_atrw_256x256-546c4594_20210414.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_atrw_256x256_20210414.log.json) |
+| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res101_atrw_256x256.py) | 256x256 | 0.898 | 0.973 | 0.936 | 0.927 | 0.985 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_atrw_256x256-da93f371_20210414.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_atrw_256x256_20210414.log.json) |
+| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res152_atrw_256x256.py) | 256x256 | 0.896 | 0.973 | 0.931 | 0.927 | 0.985 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_atrw_256x256-2bb8e162_20210414.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_atrw_256x256_20210414.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/resnet_atrw.yml

@@ -0,0 +1,56 @@
+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res50_atrw_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: ATRW
+  Name: topdown_heatmap_res50_atrw_256x256
+  Results:
+  - Dataset: ATRW
+    Metrics:
+      AP: 0.9
+      AP@0.5: 0.973
+      AP@0.75: 0.932
+      AR: 0.929
+      AR@0.5: 0.985
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_atrw_256x256-546c4594_20210414.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res101_atrw_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: ATRW
+  Name: topdown_heatmap_res101_atrw_256x256
+  Results:
+  - Dataset: ATRW
+    Metrics:
+      AP: 0.898
+      AP@0.5: 0.973
+      AP@0.75: 0.936
+      AR: 0.927
+      AR@0.5: 0.985
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_atrw_256x256-da93f371_20210414.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/atrw/res152_atrw_256x256.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: ATRW
+  Name: topdown_heatmap_res152_atrw_256x256
+  Results:
+  - Dataset: ATRW
+    Metrics:
+      AP: 0.896
+      AP@0.5: 0.973
+      AP@0.75: 0.931
+      AR: 0.927
+      AR@0.5: 0.985
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_atrw_256x256-2bb8e162_20210414.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res101_fly_192x192.py

@@ -0,0 +1,130 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/fly.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=32,
+    dataset_joints=32,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[192, 192],
+    heatmap_size=[48, 48],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=['image_file', 'center', 'scale', 'rotation', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/fly'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res152_fly_192x192.py

@@ -0,0 +1,130 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/fly.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=32,
+    dataset_joints=32,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[192, 192],
+    heatmap_size=[48, 48],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=['image_file', 'center', 'scale', 'rotation', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/fly'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res50_fly_192x192.py

@@ -0,0 +1,130 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/fly.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=32,
+    dataset_joints=32,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[192, 192],
+    heatmap_size=[48, 48],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=['image_file', 'center', 'scale', 'rotation', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/fly'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalFlyDataset',
+        ann_file=f'{data_root}/annotations/fly_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/resnet_fly.md

@@ -0,0 +1,44 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://www.nature.com/articles/s41592-018-0234-5">Vinegar Fly (Nature Methods'2019)</a></summary>
+
+```bibtex
+@article{pereira2019fast,
+  title={Fast animal pose estimation using deep neural networks},
+  author={Pereira, Talmo D and Aldarondo, Diego E and Willmore, Lindsay and Kislin, Mikhail and Wang, Samuel S-H and Murthy, Mala and Shaevitz, Joshua W},
+  journal={Nature methods},
+  volume={16},
+  number={1},
+  pages={117--125},
+  year={2019},
+  publisher={Nature Publishing Group}
+}
+```
+
+</details>
+
+Results on Vinegar Fly test set
+
+|  Arch  | Input Size | PCK@0.2 |  AUC  |  EPE  | ckpt    | log     |
+| :-------- | :--------: | :------: | :------: | :------: |:------: |:------: |
+|[pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res50_fly_192x192.py) | 192x192 | 0.996 | 0.910 | 2.00 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_fly_192x192-5d0ee2d9_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_fly_192x192_20210407.log.json) |
+|[pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res101_fly_192x192.py) | 192x192 | 0.996 | 0.912 | 1.95 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_fly_192x192-41a7a6cc_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_fly_192x192_20210407.log.json) |
+|[pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res152_fly_192x192.py) | 192x192 | 0.997 | 0.917 | 1.78 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_fly_192x192-fcafbd5a_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_fly_192x192_20210407.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/resnet_fly.yml

@@ -0,0 +1,50 @@
+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res50_fly_192x192.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: Vinegar Fly
+  Name: topdown_heatmap_res50_fly_192x192
+  Results:
+  - Dataset: Vinegar Fly
+    Metrics:
+      AUC: 0.91
+      EPE: 2.0
+      PCK@0.2: 0.996
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_fly_192x192-5d0ee2d9_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res101_fly_192x192.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Vinegar Fly
+  Name: topdown_heatmap_res101_fly_192x192
+  Results:
+  - Dataset: Vinegar Fly
+    Metrics:
+      AUC: 0.912
+      EPE: 1.95
+      PCK@0.2: 0.996
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_fly_192x192-41a7a6cc_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/fly/res152_fly_192x192.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Vinegar Fly
+  Name: topdown_heatmap_res152_fly_192x192
+  Results:
+  - Dataset: Vinegar Fly
+    Metrics:
+      AUC: 0.917
+      EPE: 1.78
+      PCK@0.2: 0.997
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_fly_192x192-fcafbd5a_20210407.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_horse10.md

@@ -0,0 +1,44 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html">HRNet (CVPR'2019)</a></summary>
+
+```bibtex
+@inproceedings{sun2019deep,
+  title={Deep high-resolution representation learning for human pose estimation},
+  author={Sun, Ke and Xiao, Bin and Liu, Dong and Wang, Jingdong},
+  booktitle={Proceedings of the IEEE conference on computer vision and pattern recognition},
+  pages={5693--5703},
+  year={2019}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://openaccess.thecvf.com/content/WACV2021/html/Mathis_Pretraining_Boosts_Out-of-Domain_Robustness_for_Pose_Estimation_WACV_2021_paper.html">Horse-10 (WACV'2021)</a></summary>
+
+```bibtex
+@inproceedings{mathis2021pretraining,
+  title={Pretraining boosts out-of-domain robustness for pose estimation},
+  author={Mathis, Alexander and Biasi, Thomas and Schneider, Steffen and Yuksekgonul, Mert and Rogers, Byron and Bethge, Matthias and Mathis, Mackenzie W},
+  booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
+  pages={1859--1868},
+  year={2021}
+}
+```
+
+</details>
+
+Results on Horse-10 test set
+
+|Set   | Arch  | Input Size | PCK@0.3 |  NME  | ckpt    | log     |
+| :--- | :---: | :--------: | :------: | :------: |:------: |:------: |
+|split1| [pose_hrnet_w32](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split1.py) | 256x256 | 0.951 | 0.122 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split1-401d901a_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split1_20210405.log.json) |
+|split2| [pose_hrnet_w32](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split2.py) | 256x256 | 0.949 | 0.116 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split2-04840523_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split2_20210405.log.json) |
+|split3| [pose_hrnet_w32](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split3.py) | 256x256 | 0.939 | 0.153 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split3-4db47400_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split3_20210405.log.json) |
+|split1| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split1.py) | 256x256 | 0.973 | 0.095 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split1-3c950d3b_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split1_20210405.log.json) |
+|split2| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split2.py) | 256x256 | 0.969 | 0.101 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split2-8ef72b5d_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split2_20210405.log.json) |
+|split3| [pose_hrnet_w48](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split3.py) | 256x256 | 0.961 | 0.128 | [ckpt](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split3-0232ec47_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split3_20210405.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_horse10.yml

@@ -0,0 +1,86 @@
+Collections:
+- Name: HRNet
+  Paper:
+    Title: Deep high-resolution representation learning for human pose estimation
+    URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Deep_High-Resolution_Representation_Learning_for_Human_Pose_Estimation_CVPR_2019_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/backbones/hrnet.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split1.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: &id001
+    - HRNet
+    Training Data: Horse-10
+  Name: topdown_heatmap_hrnet_w32_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.122
+      PCK@0.3: 0.951
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split1-401d901a_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split2.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_hrnet_w32_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.116
+      PCK@0.3: 0.949
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split2-04840523_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split3.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_hrnet_w32_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.153
+      PCK@0.3: 0.939
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w32_horse10_256x256_split3-4db47400_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split1.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_hrnet_w48_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.095
+      PCK@0.3: 0.973
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split1-3c950d3b_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split2.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_hrnet_w48_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.101
+      PCK@0.3: 0.969
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split2-8ef72b5d_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split3.py
+  In Collection: HRNet
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_hrnet_w48_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.128
+      PCK@0.3: 0.961
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/hrnet/hrnet_w48_horse10_256x256_split3-0232ec47_20210405.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split1.py

@@ -0,0 +1,164 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split2.py

@@ -0,0 +1,164 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w32_horse10_256x256-split3.py

@@ -0,0 +1,164 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w32-36af842e.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(32, 64)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(32, 64, 128)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(32, 64, 128, 256))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=32,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split1.py

@@ -0,0 +1,164 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w48-8ef0771d.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(48, 96)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(48, 96, 192)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(48, 96, 192, 384))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split2.py

@@ -0,0 +1,164 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w48-8ef0771d.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(48, 96)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(48, 96, 192)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(48, 96, 192, 384))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/hrnet_w48_horse10_256x256-split3.py

@@ -0,0 +1,164 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='https://download.openmmlab.com/mmpose/'
+    'pretrain_models/hrnet_w48-8ef0771d.pth',
+    backbone=dict(
+        type='HRNet',
+        in_channels=3,
+        extra=dict(
+            stage1=dict(
+                num_modules=1,
+                num_branches=1,
+                block='BOTTLENECK',
+                num_blocks=(4, ),
+                num_channels=(64, )),
+            stage2=dict(
+                num_modules=1,
+                num_branches=2,
+                block='BASIC',
+                num_blocks=(4, 4),
+                num_channels=(48, 96)),
+            stage3=dict(
+                num_modules=4,
+                num_branches=3,
+                block='BASIC',
+                num_blocks=(4, 4, 4),
+                num_channels=(48, 96, 192)),
+            stage4=dict(
+                num_modules=3,
+                num_branches=4,
+                block='BASIC',
+                num_blocks=(4, 4, 4, 4),
+                num_channels=(48, 96, 192, 384))),
+    ),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=48,
+        out_channels=channel_cfg['num_output_channels'],
+        num_deconv_layers=0,
+        extra=dict(final_conv_kernel=1, ),
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split1.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split2.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split3.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split1.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split2.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split3.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split1.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split1.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split2.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split2.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split3.py

@@ -0,0 +1,133 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/horse10.py'
+]
+evaluation = dict(interval=10, metric='PCK', save_best='PCK')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=22,
+    dataset_joints=22,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 21
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        21
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[256, 256],
+    heatmap_size=[64, 64],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=40, scale_factor=0.5),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'bbox_score', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=[
+            'image_file', 'center', 'scale', 'rotation', 'bbox_score',
+            'flip_pairs'
+        ]),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/horse10'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-train-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalHorse10Dataset',
+        ann_file=f'{data_root}/annotations/horse10-test-split3.json',
+        img_prefix=f'{data_root}/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/resnet_horse10.md

@@ -0,0 +1,47 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://openaccess.thecvf.com/content/WACV2021/html/Mathis_Pretraining_Boosts_Out-of-Domain_Robustness_for_Pose_Estimation_WACV_2021_paper.html">Horse-10 (WACV'2021)</a></summary>
+
+```bibtex
+@inproceedings{mathis2021pretraining,
+  title={Pretraining boosts out-of-domain robustness for pose estimation},
+  author={Mathis, Alexander and Biasi, Thomas and Schneider, Steffen and Yuksekgonul, Mert and Rogers, Byron and Bethge, Matthias and Mathis, Mackenzie W},
+  booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
+  pages={1859--1868},
+  year={2021}
+}
+```
+
+</details>
+
+Results on Horse-10 test set
+
+|Set   | Arch  | Input Size | PCK@0.3 |  NME  | ckpt    | log     |
+| :--- | :---: | :--------: | :------: | :------: |:------: |:------: |
+|split1| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split1.py) | 256x256 | 0.956 | 0.113 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split1-3a3dc37e_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split1_20210405.log.json) |
+|split2| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split2.py) | 256x256 | 0.954 | 0.111 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split2-65e2a508_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split2_20210405.log.json) |
+|split3| [pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split3.py) | 256x256 | 0.946 | 0.129 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split3-9637d4eb_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split3_20210405.log.json) |
+|split1| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split1.py) | 256x256 | 0.958 | 0.115 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split1-1b7c259c_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split1_20210405.log.json) |
+|split2| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split2.py) | 256x256 | 0.955 | 0.115 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split2-30e2fa87_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split2_20210405.log.json) |
+|split3| [pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split3.py) | 256x256 | 0.946 | 0.126 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split3-2eea5bb1_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split3_20210405.log.json) |
+|split1| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split1.py) | 256x256 | 0.969 | 0.105 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split1-7e81fe2d_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split1_20210405.log.json) |
+|split2| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split2.py) | 256x256 | 0.970 | 0.103 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split2-3b3404a3_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split2_20210405.log.json) |
+|split3| [pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split3.py) | 256x256 | 0.957 | 0.131 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split3-c957dac5_20210405.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split3_20210405.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/resnet_horse10.yml

@@ -0,0 +1,125 @@
+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split1.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: Horse-10
+  Name: topdown_heatmap_res50_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.113
+      PCK@0.3: 0.956
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split1-3a3dc37e_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split2.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res50_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.111
+      PCK@0.3: 0.954
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split2-65e2a508_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res50_horse10_256x256-split3.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res50_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.129
+      PCK@0.3: 0.946
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_horse10_256x256_split3-9637d4eb_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split1.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res101_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.115
+      PCK@0.3: 0.958
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split1-1b7c259c_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split2.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res101_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.115
+      PCK@0.3: 0.955
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split2-30e2fa87_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res101_horse10_256x256-split3.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res101_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.126
+      PCK@0.3: 0.946
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_horse10_256x256_split3-2eea5bb1_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split1.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res152_horse10_256x256-split1
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.105
+      PCK@0.3: 0.969
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split1-7e81fe2d_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split2.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res152_horse10_256x256-split2
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.103
+      PCK@0.3: 0.97
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split2-3b3404a3_20210405.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/horse10/res152_horse10_256x256-split3.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Horse-10
+  Name: topdown_heatmap_res152_horse10_256x256-split3
+  Results:
+  - Dataset: Horse-10
+    Metrics:
+      NME: 0.131
+      PCK@0.3: 0.957
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_horse10_256x256_split3-c957dac5_20210405.pth

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res101_locust_160x160.py

@@ -0,0 +1,130 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/locust.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=35,
+    dataset_joints=35,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet101',
+    backbone=dict(type='ResNet', depth=101),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[160, 160],
+    heatmap_size=[40, 40],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=['image_file', 'center', 'scale', 'rotation', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/locust'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res152_locust_160x160.py

@@ -0,0 +1,130 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/locust.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=35,
+    dataset_joints=35,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet152',
+    backbone=dict(type='ResNet', depth=152),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[160, 160],
+    heatmap_size=[40, 40],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=['image_file', 'center', 'scale', 'rotation', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/locust'
+data = dict(
+    samples_per_gpu=32,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=test_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res50_locust_160x160.py

@@ -0,0 +1,130 @@
+_base_ = [
+    '../../../../_base_/default_runtime.py',
+    '../../../../_base_/datasets/locust.py'
+]
+evaluation = dict(interval=10, metric=['PCK', 'AUC', 'EPE'], save_best='AUC')
+
+optimizer = dict(
+    type='Adam',
+    lr=5e-4,
+)
+optimizer_config = dict(grad_clip=None)
+# learning policy
+lr_config = dict(
+    policy='step',
+    warmup='linear',
+    warmup_iters=500,
+    warmup_ratio=0.001,
+    step=[170, 200])
+total_epochs = 210
+log_config = dict(
+    interval=1,
+    hooks=[
+        dict(type='TextLoggerHook'),
+        # dict(type='TensorboardLoggerHook')
+    ])
+
+channel_cfg = dict(
+    num_output_channels=35,
+    dataset_joints=35,
+    dataset_channel=[
+        [
+            0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18,
+            19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+        ],
+    ],
+    inference_channel=[
+        0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19,
+        20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34
+    ])
+
+# model settings
+model = dict(
+    type='TopDown',
+    pretrained='torchvision://resnet50',
+    backbone=dict(type='ResNet', depth=50),
+    keypoint_head=dict(
+        type='TopdownHeatmapSimpleHead',
+        in_channels=2048,
+        out_channels=channel_cfg['num_output_channels'],
+        loss_keypoint=dict(type='JointsMSELoss', use_target_weight=True)),
+    train_cfg=dict(),
+    test_cfg=dict(
+        flip_test=True,
+        post_process='default',
+        shift_heatmap=True,
+        modulate_kernel=11))
+
+data_cfg = dict(
+    image_size=[160, 160],
+    heatmap_size=[40, 40],
+    num_output_channels=channel_cfg['num_output_channels'],
+    num_joints=channel_cfg['dataset_joints'],
+    dataset_channel=channel_cfg['dataset_channel'],
+    inference_channel=channel_cfg['inference_channel'])
+
+train_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownRandomFlip', flip_prob=0.5),
+    dict(
+        type='TopDownGetRandomScaleRotation', rot_factor=90, scale_factor=0.3),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(type='TopDownGenerateTarget', sigma=2),
+    dict(
+        type='Collect',
+        keys=['img', 'target', 'target_weight'],
+        meta_keys=[
+            'image_file', 'joints_3d', 'joints_3d_visible', 'center', 'scale',
+            'rotation', 'flip_pairs'
+        ]),
+]
+
+val_pipeline = [
+    dict(type='LoadImageFromFile'),
+    dict(type='TopDownAffine'),
+    dict(type='ToTensor'),
+    dict(
+        type='NormalizeTensor',
+        mean=[0.485, 0.456, 0.406],
+        std=[0.229, 0.224, 0.225]),
+    dict(
+        type='Collect',
+        keys=['img'],
+        meta_keys=['image_file', 'center', 'scale', 'rotation', 'flip_pairs']),
+]
+
+test_pipeline = val_pipeline
+
+data_root = 'data/locust'
+data = dict(
+    samples_per_gpu=64,
+    workers_per_gpu=2,
+    val_dataloader=dict(samples_per_gpu=32),
+    test_dataloader=dict(samples_per_gpu=32),
+    train=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_train.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=train_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    val=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+    test=dict(
+        type='AnimalLocustDataset',
+        ann_file=f'{data_root}/annotations/locust_test.json',
+        img_prefix=f'{data_root}/images/',
+        data_cfg=data_cfg,
+        pipeline=val_pipeline,
+        dataset_info={{_base_.dataset_info}}),
+)

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/resnet_locust.md

@@ -0,0 +1,43 @@
+<!-- [ALGORITHM] -->
+
+<details>
+<summary align="right"><a href="http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html">SimpleBaseline2D (ECCV'2018)</a></summary>
+
+```bibtex
+@inproceedings{xiao2018simple,
+  title={Simple baselines for human pose estimation and tracking},
+  author={Xiao, Bin and Wu, Haiping and Wei, Yichen},
+  booktitle={Proceedings of the European conference on computer vision (ECCV)},
+  pages={466--481},
+  year={2018}
+}
+```
+
+</details>
+
+<!-- [DATASET] -->
+
+<details>
+<summary align="right"><a href="https://elifesciences.org/articles/47994">Desert Locust (Elife'2019)</a></summary>
+
+```bibtex
+@article{graving2019deepposekit,
+  title={DeepPoseKit, a software toolkit for fast and robust animal pose estimation using deep learning},
+  author={Graving, Jacob M and Chae, Daniel and Naik, Hemal and Li, Liang and Koger, Benjamin and Costelloe, Blair R and Couzin, Iain D},
+  journal={Elife},
+  volume={8},
+  pages={e47994},
+  year={2019},
+  publisher={eLife Sciences Publications Limited}
+}
+```
+
+</details>
+
+Results on Desert Locust test set
+
+|  Arch  | Input Size | PCK@0.2 |  AUC  |  EPE  | ckpt    | log     |
+| :-------- | :--------: | :------: | :------: | :------: |:------: |:------: |
+|[pose_resnet_50](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res50_locust_160x160.py) | 160x160 | 0.999 | 0.899 | 2.27 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res50_locust_160x160-9efca22b_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res50_locust_160x160_20210407.log.json) |
+|[pose_resnet_101](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res101_locust_160x160.py) | 160x160 | 0.999 | 0.907 | 2.03 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res101_locust_160x160-d77986b3_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res101_locust_160x160_20210407.log.json) |
+|[pose_resnet_152](/configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res152_locust_160x160.py) | 160x160 | 1.000 | 0.926 | 1.48 | [ckpt](https://download.openmmlab.com/mmpose/animal/resnet/res152_locust_160x160-4ea9b372_20210407.pth) | [log](https://download.openmmlab.com/mmpose/animal/resnet/res152_locust_160x160_20210407.log.json) |

configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/resnet_locust.yml

@@ -0,0 +1,50 @@
+Collections:
+- Name: SimpleBaseline2D
+  Paper:
+    Title: Simple baselines for human pose estimation and tracking
+    URL: http://openaccess.thecvf.com/content_ECCV_2018/html/Bin_Xiao_Simple_Baselines_for_ECCV_2018_paper.html
+  README: https://github.com/open-mmlab/mmpose/blob/master/docs/en/papers/algorithms/simplebaseline2d.md
+Models:
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res50_locust_160x160.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: &id001
+    - SimpleBaseline2D
+    Training Data: Desert Locust
+  Name: topdown_heatmap_res50_locust_160x160
+  Results:
+  - Dataset: Desert Locust
+    Metrics:
+      AUC: 0.899
+      EPE: 2.27
+      PCK@0.2: 0.999
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res50_locust_160x160-9efca22b_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res101_locust_160x160.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Desert Locust
+  Name: topdown_heatmap_res101_locust_160x160
+  Results:
+  - Dataset: Desert Locust
+    Metrics:
+      AUC: 0.907
+      EPE: 2.03
+      PCK@0.2: 0.999
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res101_locust_160x160-d77986b3_20210407.pth
+- Config: configs/animal/2d_kpt_sview_rgb_img/topdown_heatmap/locust/res152_locust_160x160.py
+  In Collection: SimpleBaseline2D
+  Metadata:
+    Architecture: *id001
+    Training Data: Desert Locust
+  Name: topdown_heatmap_res152_locust_160x160
+  Results:
+  - Dataset: Desert Locust
+    Metrics:
+      AUC: 0.926
+      EPE: 1.48
+      PCK@0.2: 1.0
+    Task: Animal 2D Keypoint
+  Weights: https://download.openmmlab.com/mmpose/animal/resnet/res152_locust_160x160-4ea9b372_20210407.pth