🤖 AI Summary
This work addresses the realistic yet underexplored challenge in person re-identification (ReID) where pedestrians simultaneously undergo clothing changes and body occlusions. We formally define this problem as Occluded Clothing-Changing ReID (OC4-ReID) and introduce the first dedicated benchmark datasets: Occ-LTCC and Occ-PRCC. To tackle occlusion-induced feature degradation and cross-clothing representation inconsistency, we propose three key components: (1) a Train-Test Micro-granularity Screening (T2MGS) module for fine-grained sample selection between training and testing phases; (2) a Part-Robust Triplet (PRT) loss to enhance local feature robustness against occlusion and appearance variation; and (3) an occlusion modeling mechanism coupled with cross-domain occlusion-robust representation learning. Extensive experiments demonstrate that our method achieves significant improvements over state-of-the-art methods on both Occ-LTCC and Occ-PRCC, while maintaining superior performance on the original CC-ReID benchmarks (LTCC and PRCC).
📝 Abstract
The study of Cloth-Changing Person Re-identification (CC-ReID) focuses on retrieving specific pedestrians when their clothing has changed, typically under the assumption that the entire pedestrian images are visible. Pedestrian images in real-world scenarios, however, are often partially obscured by obstacles, presenting a significant challenge to existing CC-ReID systems. In this paper, we introduce a more challenging task termed Occluded Cloth-Changing Person Re-Identification (OC4-ReID), which simultaneously addresses two challenges of clothing changes and occlusion. Concretely, we construct two new datasets, Occ-LTCC and Occ-PRCC, based on original CC-ReID datasets to include random occlusions of key pedestrians components (e.g., head, torso). Moreover, a novel benchmark is proposed for OC4-ReID incorporating a Train-Test Micro Granularity Screening (T2MGS) module to mitigate the influence of occlusion and proposing a Part-Robust Triplet (PRT) loss for partial features learning. Comprehensive experiments on the proposed datasets, as well as on two CC-ReID benchmark datasets demonstrate the superior performance of proposed method against other state-of-the-art methods. The codes and datasets are available at: https://github.com/1024AILab/OC4-ReID.