Self-Reinforcing Prototype Evolution with Dual-Knowledge Cooperation for Semi-Supervised Lifelong Person Re-Identification

2025-07-02

AI Summary

Semi-supervised lifelong person re-identification (Semi-LReID) faces severe challenges in real-world scenarios where labeled data are scarce yet unlabeled data are abundant; existing methods suffer from performance degradation over time due to noisy pseudo-labels in unlabeled data. Method: We propose a dynamic prototype-guided self-enhancement framework: (i) learnable identity prototypes model inter-class distributions; (ii) a dual knowledge purification mechanism jointly leverages old and new models; and (iii) dynamic clustering is integrated with pseudo-label refinement to suppress label noise and calibrate the feature space. Contribution/Results: We establish the first standardized evaluation protocol for Semi-LReID. Extensive experiments on multiple benchmarks demonstrate significant improvements over state-of-the-art methods, validating the framework’s effectiveness in enabling stable knowledge accumulation and continuous model evolution throughout lifelong learning.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Computer Vision: Multi-modal Vision
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems

Abstract

Current lifelong person re-identification (LReID) methods predominantly rely on fully labeled data streams. However, in real-world scenarios where annotation resources are limited, a vast amount of unlabeled data coexists with scarce labeled samples, leading to the Semi-Supervised LReID (Semi-LReID) problem where LReID methods suffer severe performance degradation. Existing LReID methods, even when combined with semi-supervised strategies, suffer from limited long-term adaptation performance due to struggling with the noisy knowledge occurring during unlabeled data utilization. In this paper, we pioneer the investigation of Semi-LReID, introducing a novel Self-Reinforcing Prototype Evolution with Dual-Knowledge Cooperation framework (SPRED). Our key innovation lies in establishing a self-reinforcing cycle between dynamic prototype-guided pseudo-label generation and new-old knowledge collaborative purification to enhance the utilization of unlabeled data. Specifically, learnable identity prototypes are introduced to dynamically capture the identity distributions and generate high-quality pseudo-labels. Then, the dual-knowledge cooperation scheme integrates current model specialization and historical model generalization, refining noisy pseudo-labels. Through this cyclic design, reliable pseudo-labels are progressively mined to improve current-stage learning and ensure positive knowledge propagation over long-term learning. Experiments on the established Semi-LReID benchmarks show that our SPRED achieves state-of-the-art performance. Our source code is available at https://github.com/zhoujiahuan1991/ICCV2025-SPRED

Problem

• Semi-supervised lifelong person re-identification with limited labeled data
• Noisy knowledge hinders long-term adaptation in unlabeled data utilization
• Dynamic prototype-guided pseudo-label generation and knowledge purification

Innovation

• Self-reinforcing cycle for pseudo-label generation
• Dynamic prototype-guided identity distribution capture
• Dual-knowledge cooperation for noise purification