Learning to Disentangle Scenes for Person Re-identification

TL;DR

This paper introduces a multi-branch network with self-supervision operations to disentangle complex scenes in person re-identification, significantly improving performance on multiple benchmarks by handling occlusion and scale variation more effectively.

Contribution

It proposes a divide-and-conquer approach with a novel multi-branch network and self-supervision operations to better handle scene complexity in person ReID.

Findings

Achieves state-of-the-art results on three ReID benchmarks.

Significantly improves performance on occluded ReID datasets.

Self-supervision operations enhance robustness across various challenging scenes.

Abstract

There are many challenging problems in the person re-identification (ReID) task, such as the occlusion and scale variation. Existing works usually tried to solve them by employing a one-branch network. This one-branch network needs to be robust to various challenging problems, which makes this network overburdened. This paper proposes to divide-and-conquer the ReID task. For this purpose, we employ several self-supervision operations to simulate different challenging problems and handle each challenging problem using different networks. Concretely, we use the random erasing operation and propose a novel random scaling operation to generate new images with controllable characteristics. A general multi-branch network, including one master branch and two servant branches, is introduced to handle different scenes. These branches learn collaboratively and achieve different perceptive abilities. In this way, the complex scenes in the ReID task are effectively disentangled, and the burden of each branch is relieved. The results from extensive experiments demonstrate that the proposed method achieves state-of-the-art performances on three ReID benchmarks and two occluded ReID benchmarks. Ablation study also shows that the proposed scheme and operations significantly improve the performance in various scenes. The code is available at https://git.openi.org.cn/zangxh/LDS.git.