fhw
/
ViTPose
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
32 Commits
2 Branches
0 Tags
59 MiB
 
Tree: d4deb7eef1
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from 'd4deb7eef1'
${ noResults }
ViTPose/configs/body/2d_kpt_sview_rgb_img/topdown_heatmap
Yufei 4fd8507ad3
update training params for vitpose-s 		3 years ago
..
aic bug fix in MoEMlp input params 3 years ago
coco update training params for vitpose-s 3 years ago
crowdpose register vitpose-b/l/h for coco 3 years ago
h36m register vitpose-b/l/h for coco 3 years ago
jhmdb register vitpose-b/l/h for coco 3 years ago
mhp register vitpose-b/l/h for coco 3 years ago
mpii bug fix in MoEMlp input params 3 years ago
mpii_trb register vitpose-b/l/h for coco 3 years ago
ochuman bug fix in MoEMlp input params 3 years ago
posetrack18 register vitpose-b/l/h for coco 3 years ago
README.md register vitpose-b/l/h for coco 3 years ago

README.md

Top-down heatmap-based pose estimation

Top-down methods divide the task into two stages: human detection and pose estimation.

They perform human detection first, followed by single-person pose estimation given human bounding boxes. Instead of estimating keypoint coordinates directly, the pose estimator will produce heatmaps which represent the likelihood of being a keypoint.

Various neural network models have been proposed for better performance. The popular ones include stacked hourglass networks, and HRNet.