Domain Generalization for Person Re-identification: A Survey Towards Domain-Agnostic Person Matching

2025-06-14

AI Summary

Person re-identification (ReID) suffers from significant domain shifts across camera views, and domain-generalizable ReID (DG-ReID) aims to learn domain-invariant yet identity-discriminative features without accessing target-domain data—offering greater practical relevance but remaining underexplored. This paper presents the first systematic taxonomy for DG-ReID, unifying methodological advances along three axes: multi-source input configurations, domain-generalization module design, and cross-task transfer. It comprehensively covers key techniques—including backbone optimization, feature alignment, meta-learning, data augmentation, style transfer, and self-supervised representation learning. Extensive evaluation on benchmarks such as Market-1501 and DukeMTMC demonstrates consistent improvements in cross-domain robustness. The survey further identifies critical open challenges—including scalability to unseen domains, theoretical foundations of domain invariance, and integration with foundation models—and outlines promising future research directions.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Adversarial Attacks & Robustness
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Person Re-identification (ReID) aims to retrieve images of the same individual captured across non-overlapping camera views, making it a critical component of intelligent surveillance systems. Traditional ReID methods assume that the training and test domains share similar characteristics and primarily focus on learning discriminative features within a given domain. However, they often fail to generalize to unseen domains due to domain shifts caused by variations in viewpoint, background, and lighting conditions. To address this issue, Domain-Adaptive ReID (DA-ReID) methods have been proposed. These approaches incorporate unlabeled target domain data during training and improve performance by aligning feature distributions between source and target domains. Domain-Generalizable ReID (DG-ReID) tackles a more realistic and challenging setting by aiming to learn domain-invariant features without relying on any target domain data. Recent methods have explored various strategies to enhance generalization across diverse environments, but the field remains relatively underexplored. In this paper, we present a comprehensive survey of DG-ReID. We first review the architectural components of DG-ReID including the overall setting, commonly used backbone networks and multi-source input configurations. Then, we categorize and analyze domain generalization modules that explicitly aim to learn domain-invariant and identity-discriminative representations. To examine the broader applicability of these techniques, we further conduct a case study on a related task that also involves distribution shifts. Finally, we discuss recent trends, open challenges, and promising directions for future research in DG-ReID. To the best of our knowledge, this is the first systematic survey dedicated to DG-ReID.

Problem

• Addressing domain shifts in Person Re-identification across diverse environments
• Learning domain-invariant features without target domain data
• Surveying methods for Domain-Generalizable ReID and future challenges

Innovation

• Domain-Adaptive ReID aligns feature distributions
• Domain-Generalizable ReID learns invariant features
• Survey categorizes domain generalization modules