Generalized Source-free Domain Adaptation

TL;DR

This paper introduces G-SFDA, a novel domain adaptation approach that maintains source performance while adapting to unlabeled target data, using local structure clustering and sparse domain attention.

Contribution

It proposes a new paradigm for source-free domain adaptation that preserves source accuracy and adapts effectively to target domains using innovative clustering and attention mechanisms.

Findings

Achieves state-of-the-art 85.4% on VisDA.

Performs well on multiple target domains.

Maintains source performance during adaptation.

Abstract

Domain adaptation (DA) aims to transfer the knowledge learned from a source domain to an unlabeled target domain. Some recent works tackle source-free domain adaptation (SFDA) where only a source pre-trained model is available for adaptation to the target domain. However, those methods do not consider keeping source performance which is of high practical value in real world applications. In this paper, we propose a new domain adaptation paradigm called Generalized Source-free Domain Adaptation (G-SFDA), where the learned model needs to perform well on both the target and source domains, with only access to current unlabeled target data during adaptation. First, we propose local structure clustering (LSC), aiming to cluster the target features with its semantically similar neighbors, which successfully adapts the model to the target domain in the absence of source data. Second, we propose sparse domain attention (SDA), it produces a binary domain specific attention to activate different feature channels for different domains, meanwhile the domain attention will be utilized to regularize the gradient during adaptation to keep source information. In the experiments, for target performance our method is on par with or better than existing DA and SFDA methods, specifically it achieves state-of-the-art performance (85.4%) on VisDA, and our method works well for all domains after adapting to single or multiple target domains. Code is available in https://github.com/Albert0147/G-SFDA.