DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation

16 Mar 2022  ·  Ailing Zeng, Xuan Ju, Lei Yang, Ruiyuan Gao, Xizhou Zhu, Bo Dai, Qiang Xu ·

This paper proposes a simple baseline framework for video-based 2D/3D human pose estimation that can achieve 10 times efficiency improvement over existing works without any performance degradation, named DeciWatch. Unlike current solutions that estimate each frame in a video, DeciWatch introduces a simple yet effective sample-denoise-recover framework that only watches sparsely sampled frames, taking advantage of the continuity of human motions and the lightweight pose representation. Specifically, DeciWatch uniformly samples less than 10% video frames for detailed estimation, denoises the estimated 2D/3D poses with an efficient Transformer architecture, and then accurately recovers the rest of the frames using another Transformer-based network. Comprehensive experimental results on three video-based human pose estimation and body mesh recovery tasks with four datasets validate the efficiency and effectiveness of DeciWatch.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
3D Human Pose Estimation 3DPW DeciWatch-PARE PA-MPJPE 46.4 # 6
MPJPE 75.5 # 8
3D Human Pose Estimation AIST++ DeciWatch MPJPE 67.2 # 2
Single-view Y # 1
3D Human Pose Estimation Human3.6M DeciWatch Average MPJPE (mm) 53.1 # 154
Pose Estimation J-HMDB DeciWatch Mean PCK@0.2 99.0 # 1
Mean PCK@0.1 94.6 # 1
Mean PCK@0.05 80.6 # 1
2D Human Pose Estimation JHMDB (2D poses only) DeciWatch PCK 98.8 # 1