Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

MSP-ReID: Hairstyle-Robust Cloth-Changing Person Re-Identification

Xiangyang He, Lin Wan

TL;DR

Hairstyle-Oriented Augmentation (HSOA) is introduced, which generates intra-identity hairstyle diversity to reduce hairstyle dependence and enhance attention to stable facial and body cues, and Cloth-Preserved Random Erasing (CPRE), which performs ratio-controlled erasing within clothing regions to suppress texture bias while retaining body shape and context.

Abstract

Cloth-Changing Person Re-Identification (CC-ReID) aims to match the same individual across cameras under varying clothing conditions. Existing approaches often remove apparel and focus on the head region to reduce clothing bias. However, treating the head holistically without distinguishing between face and hair leads to over-reliance on volatile hairstyle cues, causing performance degradation under hairstyle changes. To address this issue, we propose the Mitigating Hairstyle Distraction and Structural Preservation (MSP) framework. Specifically, MSP introduces Hairstyle-Oriented Augmentation (HSOA), which generates intra-identity hairstyle diversity to reduce hairstyle dependence and enhance attention to stable facial and body cues. To prevent the loss of structural information, we design Cloth-Preserved Random Erasing (CPRE), which performs ratio-controlled erasing within clothing regions to suppress texture bias while retaining body shape and context. Furthermore, we employ Region-based Parsing Attention (RPA) to incorporate parsing-guided priors that highlight face and limb regions while suppressing hair features. Extensive experiments on multiple CC-ReID benchmarks demonstrate that MSP achieves state-of-the-art performance, providing a robust and practical solution for long-term person re-identification.

MSP-ReID: Hairstyle-Robust Cloth-Changing Person Re-Identification

TL;DR

Hairstyle-Oriented Augmentation (HSOA) is introduced, which generates intra-identity hairstyle diversity to reduce hairstyle dependence and enhance attention to stable facial and body cues, and Cloth-Preserved Random Erasing (CPRE), which performs ratio-controlled erasing within clothing regions to suppress texture bias while retaining body shape and context.

Abstract

Cloth-Changing Person Re-Identification (CC-ReID) aims to match the same individual across cameras under varying clothing conditions. Existing approaches often remove apparel and focus on the head region to reduce clothing bias. However, treating the head holistically without distinguishing between face and hair leads to over-reliance on volatile hairstyle cues, causing performance degradation under hairstyle changes. To address this issue, we propose the Mitigating Hairstyle Distraction and Structural Preservation (MSP) framework. Specifically, MSP introduces Hairstyle-Oriented Augmentation (HSOA), which generates intra-identity hairstyle diversity to reduce hairstyle dependence and enhance attention to stable facial and body cues. To prevent the loss of structural information, we design Cloth-Preserved Random Erasing (CPRE), which performs ratio-controlled erasing within clothing regions to suppress texture bias while retaining body shape and context. Furthermore, we employ Region-based Parsing Attention (RPA) to incorporate parsing-guided priors that highlight face and limb regions while suppressing hair features. Extensive experiments on multiple CC-ReID benchmarks demonstrate that MSP achieves state-of-the-art performance, providing a robust and practical solution for long-term person re-identification.
Paper Structure (12 sections, 11 equations, 3 figures, 4 tables)

This paper contains 12 sections, 11 equations, 3 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. Problem Formulation
  4. Hairstyle-Oriented Augmentation (HSOA)
  5. Cloth-Preserved Random Erasing (CPRE)
  6. Region-based Parsing Attention (RPA)
  7. Objective
  8. Experiments
  9. Datasets and Protocols
  10. Comparison with State-of-the-Art Methods
  11. Ablation Studies
  12. Conclusion

Figures (3)

  • Figure 1: The Hairstyle Shortcut problem in CC-ReID. (a) Standard parsing merges face and hair together as "head"; (b) attention consequently focuses excessively on this region; (c) resulting models are robust to clothing changes but brittle to hairstyle variations; (d) conventional clothing erasure further removes structural cues.
  • Figure 2: Overview of MSP-ReID. HSOA (blue dashed, top-right) synthesizes same-ID different-hairstyle images. CPRE (pink dashed, bottom-right) creates raw/erased pairs with a clothing keep ratio. RPA (purple, center-top) uses parsing masks to boost face/limbs and suppress hair. Green denotes the ID branch, pink denotes the clothes branch for adversarial regularization. Inference is RGB-only using the ID branch.
  • Figure 3: Qualitative retrieval and attention. Query (top-left) and its feature map (bottom-left) are shown alongside Top-1$\sim$Top-10 retrievals. Green boxes are correct matches and red boxes are false.