Simple Baselines for Human Pose Estimation and Tracking

TL;DR

This paper introduces simple yet effective baseline methods for human pose estimation and tracking, achieving state-of-the-art results and facilitating future research in the field.

Contribution

It provides straightforward baseline approaches that simplify analysis and comparison, helping to inspire and evaluate new ideas in pose estimation and tracking.

Findings

Achieved state-of-the-art results on challenging benchmarks.

Provided open-source code for reproducibility.

Simplified baseline methods outperform complex models.

Abstract

There has been significant progress on pose estimation and increasing interests on pose tracking in recent years. At the same time, the overall algorithm and system complexity increases as well, making the algorithm analysis and comparison more difficult. This work provides simple and effective baseline methods. They are helpful for inspiring and evaluating new ideas for the field. State-of-the-art results are achieved on challenging benchmarks. The code will be available at https://github.com/leoxiaobin/pose.pytorch.