AIDA-ReID: Adaptive Intermediate Domain Adaptation for Generalizable and Source-Free Person Re-Identification

TL;DR

This paper introduces AIDA, a novel adaptive intermediate domain adaptation method for person re-identification that improves generalization and source-free performance by dynamically regulating feature mixing and regularization.

Contribution

It proposes a flexible, feedback-driven framework that synthesizes diverse intermediate representations and maintains identity consistency without requiring source data during deployment.

Findings

AIDA outperforms existing methods in domain generalization scenarios.

The framework effectively handles source-free person re-identification tasks.

Experimental results show significant accuracy improvements across multiple benchmarks.

Abstract

Person re-identification (Re-ID) aims to match images of the same individual across non-overlapping camera views and remains challenging due to domain shifts caused by variations in illumination, background, camera characteristics, and population distributions. Although supervised models perform well under matched training and testing conditions, their performance degrades significantly when deployed in unseen environments. Existing intermediate domain approaches such as IDM and IDM++ alleviate this gap by constructing bridge feature distributions between domains; however, they rely on fixed mixing strategies and joint source-target access, limiting their applicability to multi-source and source-free settings. To address these limitations, this paper proposes Adaptive Intermediate Domain Adaptation (AIDA), also referred to as Source-Free Multi-Source Intermediate Domain Adaptation (SF-MIDA). The proposed framework treats intermediate-domain learning as a dynamically regulated process, where feature mixing and regularization strength are adaptively controlled using feedback signals derived from model uncertainty and training stability. A multi-source intermediate domain generator synthesizes diverse intermediate representations, while a pseudo-mirror regularization strategy preserves identity consistency under domain perturbations. Extensive experiments across domain generalization and source-free settings demonstrate the effectiveness of the proposed framework.