DART$^3$: Leveraging Distance for Test Time Adaptation in Person Re-Identification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bhattacharya, Rajarshi, Murtaza, Shakeeb, Desrosiers, Christian, Dolz, Jose, Heritier, Maguelonne, Granger, Eric
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912392129019904
author Bhattacharya, Rajarshi
Murtaza, Shakeeb
Desrosiers, Christian
Dolz, Jose
Heritier, Maguelonne
Granger, Eric
author_facet Bhattacharya, Rajarshi
Murtaza, Shakeeb
Desrosiers, Christian
Dolz, Jose
Heritier, Maguelonne
Granger, Eric
contents 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 DART$^3$, a TTA framework specifically designed to mitigate camera-induced domain shifts in person ReID. DART$^3$ (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, DART$^3$ 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 DART$^3$ and DART$^3$ 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.
format Preprint
id arxiv_https___arxiv_org_abs_2505_18337
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DART$^3$: Leveraging Distance for Test Time Adaptation in Person Re-Identification
Bhattacharya, Rajarshi
Murtaza, Shakeeb
Desrosiers, Christian
Dolz, Jose
Heritier, Maguelonne
Granger, Eric
Computer Vision and Pattern Recognition
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 DART$^3$, a TTA framework specifically designed to mitigate camera-induced domain shifts in person ReID. DART$^3$ (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, DART$^3$ 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 DART$^3$ and DART$^3$ 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.
title DART$^3$: Leveraging Distance for Test Time Adaptation in Person Re-Identification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.18337