HPRNet: Hierarchical Point Regression for Whole-Body Human Pose Estimation

8 Jun 2021  ยท  Nermin Samet, Emre Akbas ยท

In this paper, we present a new bottom-up one-stage method for whole-body pose estimation, which we call "hierarchical point regression," or HPRNet for short. In standard body pose estimation, the locations of $\sim 17$ major joints on the human body are estimated. Differently, in whole-body pose estimation, the locations of fine-grained keypoints (68 on face, 21 on each hand and 3 on each foot) are estimated as well, which creates a scale variance problem that needs to be addressed. To handle the scale variance among different body parts, we build a hierarchical point representation of body parts and jointly regress them. The relative locations of fine-grained keypoints in each part (e.g. face) are regressed in reference to the center of that part, whose location itself is estimated relative to the person center. In addition, unlike the existing two-stage methods, our method predicts whole-body pose in a constant time independent of the number of people in an image. On the COCO WholeBody dataset, HPRNet significantly outperforms all previous bottom-up methods on the keypoint detection of all whole-body parts (i.e. body, foot, face and hand); it also achieves state-of-the-art results on face (75.4 AP) and hand (50.4 AP) keypoint detection. Code and models are available at \url{https://github.com/nerminsamet/HPRNet}.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Facial Landmark Detection COCO-WholeBody HPRNet (Hourglass-104) keypoint AP 75.4 # 1
Foot keypoint detection COCO-WholeBody HPRNet (Hourglass-104) keypoint AP 53 # 1
2D Human Pose Estimation COCO-WholeBody HPRNet WB 34.8 # 10
body 59.4 # 10
foot 53.0 # 10
face 75.4 # 7
hand 50.4 # 8
Multi-Person Pose Estimation COCO-WholeBody HPRNet (Hourglass-104) keypoint AP 59.4 # 1
Face Detection COCO-WholeBody HPRNet (DLA) AP 55.8 # 2
AP50 82.3 # 2
AP75 66.2 # 2
APM 40 # 2
APL 63.6 # 1
Face Detection COCO-WholeBody HPRNet (Hourglass-104) AP 56.4 # 1
AP50 82.4 # 1
AP75 67.1 # 1
APM 43.4 # 1
APL 63.3 # 2
Hand Pose Estimation COCO-WholeBody HPRNet (DLA) keypoint AP 47 # 2
Facial Landmark Detection COCO-WholeBody HPRNet (DLA) keypoint AP 74.6 # 2
Foot keypoint detection COCO-WholeBody HPRNet (DLA) keypoint AP 49.1 # 2
Multi-Person Pose Estimation COCO-WholeBody HPRNet (DLA) keypoint AP 55.2 # 2
Hand Pose Estimation COCO-WholeBody HPRNet (Hourglass-104) keypoint AP 50.4 # 1

Methods


No methods listed for this paper. Add relevant methods here