Unsupervised Domain Adaptation for Image Classification via Structure-Conditioned Adversarial Learning

TL;DR

This paper introduces a novel unsupervised domain adaptation method that preserves local class structures during distribution alignment using structure-conditioned adversarial learning, improving transfer effectiveness.

Contribution

It proposes a structure-conditioned adversarial learning scheme that maintains intra-class compactness, addressing limitations of global distribution alignment in UDA.

Findings

Effective in preserving local structures during domain adaptation

Improves classification accuracy in UDA scenarios

Outperforms existing global alignment methods

Abstract

Unsupervised domain adaptation (UDA) typically carries out knowledge transfer from a label-rich source domain to an unlabeled target domain by adversarial learning. In principle, existing UDA approaches mainly focus on the global distribution alignment between domains while ignoring the intrinsic local distribution properties. Motivated by this observation, we propose an end-to-end structure-conditioned adversarial learning scheme (SCAL) that is able to preserve the intra-class compactness during domain distribution alignment. By using local structures as structure-aware conditions, the proposed scheme is implemented in a structure-conditioned adversarial learning pipeline. The above learning procedure is iteratively performed by alternating between local structures establishment and structure-conditioned adversarial learning. Experimental results demonstrate the effectiveness of the proposed scheme in UDA scenarios.