CMCC-ReID: Cross-Modality Clothing-Change Person Re-Identification

TL;DR

This paper introduces a new person re-identification task addressing both clothing changes and modality differences, proposing a benchmark and a novel network to improve cross-modality and clothing-invariant matching.

Contribution

The study defines the CMCC-ReID task, creates the SYSU-CMCC benchmark, and proposes the PIA network with DBDL and BPL modules to handle clothing and modality variations.

Findings

PIA outperforms existing methods on SYSU-CMCC dataset.

The DBDL module effectively disentangles clothing from identity cues.

Bi-Directional Prototype Learning reduces modality gap and clothing interference.

Abstract

Person Re-Identification (ReID) faces severe challenges from modality discrepancy and clothing variation in long-term surveillance scenario. While existing studies have made significant progress in either Visible-Infrared ReID (VI-ReID) or Clothing-Change ReID (CC-ReID), real-world surveillance system often face both challenges simultaneously. To address this overlooked yet realistic problem, we define a new task, termed Cross-Modality Clothing-Change Re-Identification (CMCC-ReID), which targets pedestrian matching across variations in both modality and clothing. To advance research in this direction, we construct a new benchmark SYSU-CMCC, where each identity is captured in both visible and infrared domains with distinct outfits, reflecting the dual heterogeneity of long-term surveillance. To tackle CMCC-ReID, we propose a Progressive Identity Alignment Network (PIA) that progressively mitigates the issues of clothing variation and modality discrepancy. Specifically, a Dual-Branch Disentangling Learning (DBDL) module separates identity-related cues from clothing-related factors to achieve clothing-agnostic representation, and a Bi-Directional Prototype Learning (BPL) module performs intra-modality and inter-modality contrast in the embedding space to bridge the modality gap while further suppressing clothing interference. Extensive experiments on the SYSU-CMCC dataset demonstrate that PIA establishes a strong baseline for this new task and significantly outperforms existing methods.