# DART3: Leveraging Distance for Test Time Adaptation in Person Re-Identification

arXiv: https://arxiv.org/abs/2505.18337 · HTML: https://mlsanity.com/paper/2505.18337/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 25 of its 36 references that are in the map.
- First posted on arXiv 23 May 2025.

## Builds on (most cited first, 25 of 25)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking](https://mlsanity.com/paper/1609.01775/) (2016)
- [Person Transfer GAN to Bridge Domain Gap for Person Re-Identification](https://mlsanity.com/paper/1711.08565/) (2017)
- [Efficient Test-Time Model Adaptation without Forgetting](https://mlsanity.com/paper/2204.02610/) (2022)
- [TransReID: Transformer-based Object Re-Identification](https://mlsanity.com/paper/2102.04378/) (2021)
- [Contrastive Test-Time Adaptation](https://mlsanity.com/paper/2204.10377/) (2022)
- [Parameter-free Online Test-time Adaptation](https://mlsanity.com/paper/2201.05718/) (2022)
- [CLIP-ReID: Exploiting Vision-Language Model for Image Re-Identification without Concrete Text Labels](https://mlsanity.com/paper/2211.13977/) (2022)
- [Test-Time Adaptable Neural Networks for Robust Medical Image Segmentation](https://mlsanity.com/paper/2004.04668/) (2020)
- [Test-Time Adaptation via Conjugate Pseudo-labels](https://mlsanity.com/paper/2207.09640/) (2022)
- [Cloth-Changing Person Re-identification from A Single Image with Gait Prediction and Regularization](https://mlsanity.com/paper/2103.15537/) (2021)
- [ICE: Inter-instance Contrastive Encoding for Unsupervised Person Re-identification](https://mlsanity.com/paper/2103.16364/) (2021)
- [Body Part-Based Representation Learning for Occluded Person Re-Identification](https://mlsanity.com/paper/2211.03679/) (2022)
- [Camera-aware Proxies for Unsupervised Person Re-Identification](https://mlsanity.com/paper/2012.10674/) (2020)
- [Part-based Pseudo Label Refinement for Unsupervised Person Re-identification](https://mlsanity.com/paper/2203.14675/) (2022)
- [Test-time Batch Statistics Calibration for Covariate Shift](https://mlsanity.com/paper/2110.04065/) (2021)
- [DLTTA: Dynamic Learning Rate for Test-time Adaptation on Cross-domain Medical Images](https://mlsanity.com/paper/2205.13723/) (2022)
- [Test-time Adaptation with Calibration of Medical Image Classification Nets for Label Distribution Shift](https://mlsanity.com/paper/2207.00769/) (2022)
- [Self domain adapted network](https://mlsanity.com/paper/2007.03162/) (2020)
- [Camera-Driven Representation Learning for Unsupervised Domain Adaptive Person Re-identification](https://mlsanity.com/paper/2308.11901/) (2023)
- [Generalizable Person Re-Identification via Self-Supervised Batch Norm Test-Time Adaption](https://mlsanity.com/paper/2203.00672/) (2022)
- [Towards Understanding GD with Hard and Conjugate Pseudo-labels for Test-Time Adaptation](https://mlsanity.com/paper/2210.10019/) (2022)
- [1st Place Solution to VisDA-2020: Bias Elimination for Domain Adaptive Pedestrian Re-identification](https://mlsanity.com/paper/2012.13498/) (2020)
- [Test-time Similarity Modification for Person Re-identification toward Temporal Distribution Shift](https://mlsanity.com/paper/2403.14114/) (2024)
- [Exploring the Camera Bias of Person Re-identification](https://mlsanity.com/paper/2502.10195/) (2025)

## Abstract (arXiv)

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.