The Devil is in the Details: Delving into Unbiased Data Processing for Human Pose Estimation
Junjie Huang, Zheng Zhu, Feng Guo, Guan Huang, Dalong Du

TL;DR
This paper identifies biases in data processing methods like flipping and keypoint transformations that hinder human pose estimation accuracy, and proposes an unbiased data processing approach to improve performance and reproducibility.
Contribution
The paper introduces Unbiased Data Processing (UDP), a model-agnostic method addressing biases in coordinate and keypoint transformations in human pose estimation.
Findings
UDP improves pose estimation accuracy across multiple benchmarks.
Unbiased transformations lead to more reproducible and fair comparisons.
The approach sets a new reliable baseline for future research.
Abstract
Being a fundamental component in training and inference, data processing has not been systematically considered in human pose estimation community, to the best of our knowledge. In this paper, we focus on this problem and find that the devil of human pose estimation evolution is in the biased data processing. Specifically, by investigating the standard data processing in state-of-the-art approaches mainly including coordinate system transformation and keypoint format transformation (i.e., encoding and decoding), we find that the results obtained by common flipping strategy are unaligned with the original ones in inference. Moreover, there is a statistical error in some keypoint format transformation methods. Two problems couple together, significantly degrade the pose estimation performance and thus lay a trap for the research community. This trap has given bone to many suboptimal…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
The Devil Is in the Details: Delving Into Unbiased Data Processing for Human Pose Estimation· youtube
Taxonomy
TopicsHuman Pose and Action Recognition · Advanced Vision and Imaging · Human Motion and Animation
