Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regression

2021-04-06 · CVPR 2021 1 · Zigang Geng, Ke Sun, Bin Xiao, Zhaoxiang Zhang, Jingdong Wang

In this paper, we are interested in the bottom-up paradigm of estimating human poses from an image. We study the dense keypoint regression framework that is previously inferior to the keypoint detection and grouping framework. Our motivation is that regressing keypoint positions accurately needs to learn representations that focus on the keypoint regions. We present a simple yet effective approach, named disentangled keypoint regression (DEKR). We adopt adaptive convolutions through pixel-wise spatial transformer to activate the pixels in the keypoint regions and accordingly learn representations from them. We use a multi-branch structure for separate regression: each branch learns a representation with dedicated adaptive convolutions and regresses one keypoint. The resulting disentangled representations are able to attend to the keypoint regions, respectively, and thus the keypoint regression is spatially more accurate. We empirically show that the proposed direct regression method outperforms keypoint detection and grouping methods and achieves superior bottom-up pose estimation results on two benchmark datasets, COCO and CrowdPose. The code and models are available at https://github.com/HRNet/DEKR.

📄 PDF Abstract BibTeX arXiv:2104.02300

Code (2)

HRNet/DEKR 공식 구현 pytorch

open-mmlab/mmpose pytorch

Tasks

Keypoint Detection Pose Estimation

Methods 이 논문이 사용한 방법론

Spatial Transformer A Spatial Transformer is an image model block that explicitly allows the spatial manipulation of data within a [convolutional neural…

BAPose: Bottom-Up Pose Estimation with Disentangled Waterfall Representations

2021-12-20 · Bruno Artacho, Andreas Savakis

We propose BAPose, a novel bottom-up approach that achieves state-of-the-art results for multi-person pose estimation. Our end-to-end trainable framework leverages a disentangled multi-scale waterfall architecture and in…

Multi-Person Pose Estimation Pose Estimation

Bottom-Up Human Pose Estimation by Ranking Heatmap-Guided Adaptive Keypoint Estimates

2020-06-28 · Ke Sun, Zigang Geng, Depu Meng, Bin Xiao 외

The typical bottom-up human pose estimation framework includes two stages, keypoint detection and grouping. Most existing works focus on developing grouping algorithms, e.g., associative embedding, and pixel-wise keypoin…

Keypoint Detection Multi-Person Pose Estimation Pose Estimation regression

Differentiable Hierarchical Graph Grouping for Multi-Person Pose Estimation

2020-07-23 · ECCV 2020 8 · Sheng Jin, Wentao Liu, Enze Xie, Wenhai Wang 외

Multi-person pose estimation is challenging because it localizes body keypoints for multiple persons simultaneously. Previous methods can be divided into two streams, i.e. top-down and bottom-up methods. The top-down met…

2D Human Pose Estimation Clustering Graph Clustering Graph Neural Network+4

BoIR: Box-Supervised Instance Representation for Multi-Person Pose Estimation

2023-09-25 · Uyoung Jeong, Seungryul Baek, Hyung Jin Chang, Kwang In Kim

Single-stage multi-person human pose estimation (MPPE) methods have shown great performance improvements, but existing methods fail to disentangle features by individual instances under crowded scenes. In this paper, we …

Disentanglement Keypoint Estimation Multi-Person Pose Estimation Multi-Task Learning+2

2D Human Pose Estimation with Explicit Anatomical Keypoints Structure Constraints

2022-12-05 · Zhangjian Ji, Zilong Wang, Ming Zhang, Yapeng Chen 외

Recently, human pose estimation mainly focuses on how to design a more effective and better deep network structure as human features extractor, and most designed feature extraction networks only introduce the position of…

2D Human Pose Estimation Pose Estimation