2025/05/23 by Rajarshi Bhattacharya, Bhattacharya, Rajarshi, Shakeeb Murtaza +9
Computer Science · Engineering · Medicine · #Autopsy Techniques and Outcomes #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Gait Recognition and Analysis #Video Surveillance and Tracking Methods
paper · pdf · doi:10.48550/arxiv.2505.18337
openalex publication_date 2025/05/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Person re-identification (ReID) models are known to suffer from camera bias, where learned representations cluster according to camera viewpoints rather than identity, leading to significant performance degradation under (inter-camera) domain shifts in real-world surveillance systems when new cameras are added to camera networks. State-of-the-art test-time adaptation (TTA) methods, largely designed for classification tasks, rely on classification entropy-based objectives that fail to generalize well to ReID, thus making them unsuitable for tackling camera bias. In this paper, we introduce DART3, a TTA framework specifically designed to mitigate camera-induced domain shifts in person ReID. DART3 (Distance-Aware Retrieval Tuning at Test Time) leverages a distance-based objective that aligns better with image retrieval tasks like ReID by exploiting the correlation between nearest-neighbor distance and prediction error. Unlike prior ReID-specific domain adaptation methods, DART3 requires no source data, architectural modifications, or retraining, and can be deployed in both fully black-box and hybrid settings. Empirical evaluations on multiple ReID benchmarks indicate that DART3 and DART3 LITE, a lightweight alternative to the approach, consistently outperforms state-of-the-art TTA baselines, making for a viable option to online learning to mitigate the adverse effects of camera bias.