Distribution-aware Forgetting Compensation for Exemplar-Free Lifelong Person Re-identification

2025-04-21

AI Summary

Lifelong Person Re-Identification (LReID) inherently confronts the trade-off between adapting to novel domains and retaining knowledge from previously encountered domains. Existing replay-free methods suffer from progressive performance degradation due to the absence of explicit distribution modeling and forgetting mitigation. This paper proposes the first distribution-aware forgetting compensation framework that requires neither historical sample storage nor knowledge distillation. Our approach achieves cross-domain semantic alignment via text-driven prompt aggregation; jointly optimizes shared representations and domain-specific distributions through multi-expert domain distribution modeling and instance-level discriminative loss; and introduces cross-domain consistency alignment alongside high-dimensional shared-region fusion. Evaluated on two standard incremental benchmark sequences, our method achieves average improvements of 9.8%/6.6% and 6.4%/6.2% in mAP/R@1, respectively—substantially outperforming state-of-the-art approaches.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Search and Optimization: Learning to Search
• Computer Vision: Diffusion Models for Vision

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Lifelong Person Re-identification (LReID) suffers from a key challenge in preserving old knowledge while adapting to new information. The existing solutions include rehearsal-based and rehearsal-free methods to address this challenge. Rehearsal-based approaches rely on knowledge distillation, continuously accumulating forgetting during the distillation process. Rehearsal-free methods insufficiently learn the distribution of each domain, leading to forgetfulness over time. To solve these issues, we propose a novel Distribution-aware Forgetting Compensation (DAFC) model that explores cross-domain shared representation learning and domain-specific distribution integration without using old exemplars or knowledge distillation. We propose a Text-driven Prompt Aggregation (TPA) that utilizes text features to enrich prompt elements and guide the prompt model to learn fine-grained representations for each instance. This can enhance the differentiation of identity information and establish the foundation for domain distribution awareness. Then, Distribution-based Awareness and Integration (DAI) is designed to capture each domain-specific distribution by a dedicated expert network and adaptively consolidate them into a shared region in high-dimensional space. In this manner, DAI can consolidate and enhance cross-domain shared representation learning while alleviating catastrophic forgetting. Furthermore, we develop a Knowledge Consolidation Mechanism (KCM) that comprises instance-level discrimination and cross-domain consistency alignment strategies to facilitate model adaptive learning of new knowledge from the current domain and promote knowledge consolidation learning between acquired domain-specific distributions, respectively. Experimental results show that our DAFC outperform state-of-the-art methods by at least 9.8%/6.6% and 6.4%/6.2% of average mAP/R@1 on two training orders.

Problem

• Address catastrophic forgetting in lifelong person re-identification
• Enhance cross-domain shared representation learning
• Improve domain-specific distribution integration without old exemplars

Innovation

• Text-driven Prompt Aggregation for fine-grained representations
• Distribution-based Awareness and Integration for domain-specific learning
• Knowledge Consolidation Mechanism for adaptive learning