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.
7 Commits
2 Branches
0 Tags
59 MiB
 
Tree: 86eced9576
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '86eced9576'
${ noResults }
ViTPose/configs/wholebody/2d_kpt_sview_rgb_img/associative_embedding
Yufei 33f49cf1f2
register vitpose-b/l/h for coco 		3 years ago
..
coco-wholebody register vitpose-b/l/h for coco 3 years ago
README.md register vitpose-b/l/h for coco 3 years ago

README.md

Associative embedding: End-to-end learning for joint detection and grouping (AE)

Associative Embedding (NIPS'2017) 
@inproceedings{newell2017associative,
  title={Associative embedding: End-to-end learning for joint detection and grouping},
  author={Newell, Alejandro and Huang, Zhiao and Deng, Jia},
  booktitle={Advances in neural information processing systems},
  pages={2277--2287},
  year={2017}
}

AE is one of the most popular 2D bottom-up pose estimation approaches, that first detect all the keypoints and then group/associate them into person instances.

In order to group all the predicted keypoints to individuals, a tag is also predicted for each detected keypoint. Tags of the same person are similar, while tags of different people are different. Thus the keypoints can be grouped according to the tags.