articleIEEE Transactions on Pattern Analysis and Machine IntelligenceNov 17, 2022Closed access

AlphaPose: Whole-Body Regional Multi-Person Pose Estimation and Tracking in Real-Time

HFHao-Shu FangJLJiefeng LiHTHongyang TangCXChao XuHZHaoyi Zhu

Shanghai Jiao Tong University · Max Planck Institute for Intelligent Systems

Indexed incrossref

Abstract

Accurate whole-body multi-person pose estimation and tracking is an important yet challenging topic in computer vision. To capture the subtle actions of humans for complex behavior analysis, whole-body pose estimation including the face, body, hand and foot is essential over conventional body-only pose estimation. In this article, we present AlphaPose, a system that can perform accurate whole-body pose estimation and tracking jointly while running in realtime. To this end, we propose several new techniques: Symmetric Integral Keypoint Regression (SIKR) for fast and fine localization, Parametric Pose Non-Maximum-Suppression (P-NMS) for eliminating redundant human detections and Pose Aware Identity Embedding for…

Citation impact

663
total citations
FWCI
61.40
Percentile
100%
References
114
Citations per year

Authors

8

Topics & keywords

Topics
Keywords
  • Pose
  • Artificial intelligence
  • Computer science
  • Computer vision
  • Tracking (education)
  • Articulated body pose estimation
  • Estimation
  • 3D pose estimation
No related works found for this paper.

Funding