Multi-Person Pose Estimation with Local Joint-to-Person Associations

TL;DR

This paper introduces a novel multi-person pose estimation method that formulates joint-to-person association as an ILP problem, enabling accurate pose detection in crowded scenes with high efficiency.

Contribution

It proposes a local ILP-based approach for joint-to-person association in multi-person pose estimation, significantly improving speed while maintaining accuracy.

Findings

Achieves state-of-the-art accuracy on MPII dataset.

Runs 6,000 to 19,000 times faster than comparable methods.

Effectively handles occlusions and truncations in crowded scenes.

Abstract

Despite of the recent success of neural networks for human pose estimation, current approaches are limited to pose estimation of a single person and cannot handle humans in groups or crowds. In this work, we propose a method that estimates the poses of multiple persons in an image in which a person can be occluded by another person or might be truncated. To this end, we consider multi-person pose estimation as a joint-to-person association problem. We construct a fully connected graph from a set of detected joint candidates in an image and resolve the joint-to-person association and outlier detection using integer linear programming. Since solving joint-to-person association jointly for all persons in an image is an NP-hard problem and even approximations are expensive, we solve the problem locally for each person. On the challenging MPII Human Pose Dataset for multiple persons, our approach achieves the accuracy of a state-of-the-art method, but it is 6,000 to 19,000 times faster.