Cluster-guided Asymmetric Contrastive Learning for Unsupervised Person Re-Identification
This addresses the challenge of matching pedestrian images across camera views without labeled data, offering an incremental improvement over prior clustering-based methods.
The paper tackles the problem of unsupervised person re-identification, where existing methods rely on clustering that is biased by image colors, by proposing a cluster-guided asymmetric contrastive learning approach, achieving superior performance on three benchmark datasets.
Unsupervised person re-identification (Re-ID) aims to match pedestrian images from different camera views in unsupervised setting. Existing methods for unsupervised person Re-ID are usually built upon the pseudo labels from clustering. However, the quality of clustering depends heavily on the quality of the learned features, which are overwhelmingly dominated by the colors in images especially in the unsupervised setting. In this paper, we propose a Cluster-guided Asymmetric Contrastive Learning (CACL) approach for unsupervised person Re-ID, in which cluster structure is leveraged to guide the feature learning in a properly designed asymmetric contrastive learning framework. To be specific, we propose a novel cluster-level contrastive loss to help the siamese network effectively mine the invariance in feature learning with respect to the cluster structure within and between different data augmentation views, respectively. Extensive experiments conducted on three benchmark datasets demonstrate superior performance of our proposal.