Prompt-Anchored Vision-Text Distillation for Lifelong Person Re-identification
For lifelong person re-identification, this work addresses the challenge of balancing stability and plasticity in an exemplar-free setting, outperforming existing methods.
Lifelong person re-identification suffers from semantic drift and catastrophic forgetting. The proposed PAD method uses a frozen text encoder as a semantic anchor and an asymmetric vision-text distillation framework, achieving state-of-the-art performance across seen and unseen domains with a strong stability-plasticity balance.
Lifelong person re-identification (LReID) aims to train a generalizable model with sequentially collected data. However, such models often suffer from semantic drift, limited adaptability, and catastrophic forgetting as new domains emerge. Existing exemplar-free approaches largely rely on visual-only distillation or parameter regularization, while overlooking the potential of auxiliary modalities, such as text, to preserve semantic stability and enable incremental plasticity. We observe that the frozen text encoder in pretrained vision-language models can serve as a stable semantic anchor across domains. To decouple the roles of vision and text, we propose Prompt-Anchored vision-text Distillation (PAD), an asymmetric vision-text framework for semantic alignment and cross-domain generalization. On the textual side, we distill prompts to preserve vision-text alignment under a fixed semantic space, acting as a global semantic reference rather than a dominant learning signal. On the visual side, an EMA-based teacher with an adaptive prompt pool enables domain-wise adaptation by allocating new slots while freezing past ones. Extensive experiments show that PAD substantially outperforms state-of-the-art methods across seen and unseen domains, achieving a strong balance between stability and plasticity. Project page is available at https://github.com/zu-zi/PAD.