CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmark

TL;DR

This paper introduces a new efficient method and dataset for multi-person pose estimation in crowded scenes, addressing challenges not covered by previous benchmarks and surpassing existing methods in accuracy.

Contribution

A novel approach combining joint-candidate SPPE and global maximum joints association, along with a new dataset for better evaluation in crowded scenes.

Findings

Surpassed state-of-the-art on CrowdPose dataset by 5.2 mAP

Demonstrated strong generalization on MSCOCO dataset

Efficient inference suitable for crowded scenarios

Abstract

Multi-person pose estimation is fundamental to many computer vision tasks and has made significant progress in recent years. However, few previous methods explored the problem of pose estimation in crowded scenes while it remains challenging and inevitable in many scenarios. Moreover, current benchmarks cannot provide an appropriate evaluation for such cases. In this paper, we propose a novel and efficient method to tackle the problem of pose estimation in the crowd and a new dataset to better evaluate algorithms. Our model consists of two key components: joint-candidate single person pose estimation (SPPE) and global maximum joints association. With multi-peak prediction for each joint and global association using graph model, our method is robust to inevitable interference in crowded scenes and very efficient in inference. The proposed method surpasses the state-of-the-art methods on CrowdPose dataset by 5.2 mAP and results on MSCOCO dataset demonstrate the generalization ability of our method. Source code and dataset will be made publicly available.