Domain Consistency Representation Learning for Lifelong Person Re-Identification

TL;DR

This paper introduces a novel domain consistency representation learning model for lifelong person re-identification that balances intra-domain discrimination and inter-domain gaps using global and attribute-wise representations.

Contribution

The paper proposes a new DCR model that explores global and attribute-wise representations, along with an anti-forgetting strategy and knowledge consolidation for improved LReID.

Findings

DCR outperforms state-of-the-art methods in LReID tasks.

The attribute-oriented anti-forgetting strategy effectively reduces catastrophic forgetting.

Knowledge consolidation enhances inter-domain consistency.

Abstract

Lifelong person re-identification (LReID) exhibits a contradictory relationship between intra-domain discrimination and inter-domain gaps when learning from continuous data. Intra-domain discrimination focuses on individual nuances (i.e., clothing type, accessories, etc.), while inter-domain gaps emphasize domain consistency. Achieving a trade-off between maximizing intra-domain discrimination and minimizing inter-domain gaps is a crucial challenge for improving LReID performance. Most existing methods strive to reduce inter-domain gaps through knowledge distillation to maintain domain consistency. However, they often ignore intra-domain discrimination. To address this challenge, we propose a novel domain consistency representation learning (DCR) model that explores global and attribute-wise representations as a bridge to balance intra-domain discrimination and inter-domain gaps. At the intra-domain level, we explore the complementary relationship between global and attribute-wise representations to improve discrimination among similar identities. Excessive learning intra-domain discrimination can lead to catastrophic forgetting. We further develop an attribute-oriented anti-forgetting (AF) strategy that explores attribute-wise representations to enhance inter-domain consistency, and propose a knowledge consolidation (KC) strategy to facilitate knowledge transfer. Extensive experiments show that our DCR achieves superior performance compared to state-of-the-art LReID methods. Our code is available at https://github.com/LiuShiBen/DCR.