Take More Positives: An Empirical Study of Contrastive Learing in Unsupervised Person Re-Identification

TL;DR

This paper investigates the success factors of contrastive learning in unsupervised person re-identification, revealing that using more positives during training improves performance and proposing a new method that outperforms existing state-of-the-art approaches.

Contribution

It introduces a simple yet effective contrastive learning approach that leverages more positives and eliminates the need for memory back, advancing unsupervised person re-ID.

Findings

Our method outperforms state-of-the-art on benchmark datasets.

Handling hard negatives implicitly through data augmentation and sampling.

Using more positives improves unsupervised re-ID performance.

Abstract

Unsupervised person re-identification (re-ID) aims at closing the performance gap to supervised methods. These methods build reliable relationship between data points while learning representations. However, we empirically show that the reason why they are successful is not only their label generation mechanisms, but also their unexplored designs. By studying two unsupervised person re-ID methods in a cross-method way, we point out a hard negative problem is handled implicitly by their designs of data augmentations and PK sampler respectively. In this paper, we find another simple solution for the problem, i.e., taking more positives during training, by which we generate pseudo-labels and update models in an iterative manner. Based on our findings, we propose a contrastive learning method without a memory back for unsupervised person re-ID. Our method works well on benchmark datasets and outperforms the state-of-the-art methods. Code will be made available.