Towards Realistic Lifelong Re-identification: Identity Recurrence with Changing Clothes

Existing lifelong person re-identification (Re-ID) methods assume that each identity maintains a relatively stable appearance distribution over time. However, in real-world scenarios, identities often reappear asynchronously with substantial clothing changes, which is not modeled in existing lifelong Re-ID formulations. We therefore study a realistic lifelong cloth-changing Re-ID (LCC) setting, in which identities reappear asynchronously under substantial clothing changes. This setting leads to two core difficulties: the model must acquire new identities while adapting representations of recurring ones, and at the same time remain robust to the substantial representation drift induced by clothing changes over time, which undermines cross-temporal identity consistency. To address these, we develop a framework that disentangles identity-intrinsic representations from clothing-induced biases, enabling identity modeling beyond appearance changes. We further introduce a Dynamic Identity-Anchor Alignment to maintain stable identity anchors under stage-wise distribution shifts. Experiments on the LTCC and PRCC benchmarks demonstrate superior performance and representational stability across multiple learning stages. Our code is available at https://github.com/joyner-7/LCC_ReID.

In real-world scenarios, people captured by surveillance cameras often reappear at different times wearing entirely different clothes, which makes it very difficult for traditional computer vision systems to recognize them. This paper introduces a novel AI framework that successfully separates a person's permanent "identity and body structure cues" from their temporary "clothing style." By doing so, the system can stably and accurately recognize the same individual over a long period, even when they change their clothes and new data continuously flows in.