Colors See Colors Ignore: Clothes Changing ReID with Color Disentanglement

2025-07-09

AI Summary

This paper addresses the appearance discrepancy problem in clothing-changing person re-identification (CC-ReID) caused by garment variation. We propose a lightweight, annotation-free disentanglement framework that requires no auxiliary models. Our key innovation is the first use of foreground and background color statistics as unsupervised proxy signals: via color-space mapping and feature disentanglement learning, we explicitly separate color-aware features from identity-specific representations. Furthermore, we introduce a Spatial-to-Attention (S2A) self-attention mechanism to explicitly suppress information leakage between color and identity features. The end-to-end RGB network achieves significant improvements across four CC-ReID benchmarks: for image-based ReID, mAP increases by 2.9% on LTCC and 5.0% on PRCC; for video-based ReID, mAP improves by 1.0% on CCVID and 2.5% on MeVID.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Feature Construction/Reformulation
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Clothes-Changing Re-Identification (CC-ReID) aims to recognize individuals across different locations and times, irrespective of clothing. Existing methods often rely on additional models or annotations to learn robust, clothing-invariant features, making them resource-intensive. In contrast, we explore the use of color - specifically foreground and background colors - as a lightweight, annotation-free proxy for mitigating appearance bias in ReID models. We propose Colors See, Colors Ignore (CSCI), an RGB-only method that leverages color information directly from raw images or video frames. CSCI efficiently captures color-related appearance bias ('Color See') while disentangling it from identity-relevant ReID features ('Color Ignore'). To achieve this, we introduce S2A self-attention, a novel self-attention to prevent information leak between color and identity cues within the feature space. Our analysis shows a strong correspondence between learned color embeddings and clothing attributes, validating color as an effective proxy when explicit clothing labels are unavailable. We demonstrate the effectiveness of CSCI on both image and video ReID with extensive experiments on four CC-ReID datasets. We improve the baseline by Top-1 2.9% on LTCC and 5.0% on PRCC for image-based ReID, and 1.0% on CCVID and 2.5% on MeVID for video-based ReID without relying on additional supervision. Our results highlight the potential of color as a cost-effective solution for addressing appearance bias in CC-ReID. Github: https://github.com/ppriyank/ICCV-CSCI-Person-ReID.

Problem

• Mitigate appearance bias in ReID models using color
• Develop lightweight method without extra annotations
• Disentangle color-related bias from identity features

Innovation

• Uses RGB-only method for color information
• Introduces S2A self-attention for feature disentanglement
• Leverages color as proxy without clothing labels