Learning Invariance from Generated Variance for Unsupervised Person Re-identification

TL;DR

This paper introduces a GAN-based augmentation method for unsupervised person re-identification, improving invariance learning by generating diverse views that disentangle identity-related and unrelated features, leading to state-of-the-art results.

Contribution

It proposes a novel GAN-guided augmentation approach that manipulates both identity-related and unrelated features for contrastive learning in unsupervised ReID.

Findings

Achieves new state-of-the-art performance on large-scale benchmarks.

Effectively disentangles identity-related and unrelated features.

Enhances invariance learning through GAN-based augmentations.

Abstract

This work focuses on unsupervised representation learning in person re-identification (ReID). Recent self-supervised contrastive learning methods learn invariance by maximizing the representation similarity between two augmented views of a same image. However, traditional data augmentation may bring to the fore undesirable distortions on identity features, which is not always favorable in id-sensitive ReID tasks. In this paper, we propose to replace traditional data augmentation with a generative adversarial network (GAN) that is targeted to generate augmented views for contrastive learning. A 3D mesh guided person image generator is proposed to disentangle a person image into id-related and id-unrelated features. Deviating from previous GAN-based ReID methods that only work in id-unrelated space (pose and camera style), we conduct GAN-based augmentation on both id-unrelated and id-related features. We further propose specific contrastive losses to help our network learn invariance from id-unrelated and id-related augmentations. By jointly training the generative and the contrastive modules, our method achieves new state-of-the-art unsupervised person ReID performance on mainstream large-scale benchmarks.