Unsupervised Lifelong Person Re-identification via Contrastive Rehearsal

TL;DR

This paper introduces an unsupervised lifelong person re-identification approach that continuously adapts to new domains without forgetting old knowledge, using contrastive rehearsal and similarity constraints to enhance generalization across multiple domains.

Contribution

The paper proposes a novel unsupervised lifelong ReID method that maintains knowledge across domains through contrastive rehearsal and similarity regularization, outperforming previous methods.

Findings

Achieves strong generalization on seen and unseen domains.

Significantly outperforms previous lifelong ReID methods.

Effective in continuous domain adaptation without catastrophic forgetting.

Abstract

Existing unsupervised person re-identification (ReID) methods focus on adapting a model trained on a source domain to a fixed target domain. However, an adapted ReID model usually only works well on a certain target domain, but can hardly memorize the source domain knowledge and generalize to upcoming unseen data. In this paper, we propose unsupervised lifelong person ReID, which focuses on continuously conducting unsupervised domain adaptation on new domains without forgetting the knowledge learnt from old domains. To tackle unsupervised lifelong ReID, we conduct a contrastive rehearsal on a small number of stored old samples while sequentially adapting to new domains. We further set an image-to-image similarity constraint between old and new models to regularize the model updates in a way that suits old knowledge. We sequentially train our model on several large-scale datasets in an unsupervised manner and test it on all seen domains as well as several unseen domains to validate the generalizability of our method. Our proposed unsupervised lifelong method achieves strong generalizability, which significantly outperforms previous lifelong methods on both seen and unseen domains. Code will be made available at https://github.com/chenhao2345/UCR.