Transformer-Based Source-Free Domain Adaptation

TL;DR

This paper introduces TransDA, a Transformer-based framework for source-free domain adaptation that enhances model focus on object regions, improving generalization and achieving state-of-the-art results across various adaptation tasks.

Contribution

The paper proposes a novel Transformer-based approach with self-supervised knowledge distillation for improved source-free domain adaptation.

Findings

TransDA significantly outperforms existing methods in accuracy.

Attention focusing on object regions improves domain adaptation.

State-of-the-art results on multiple domain adaptation benchmarks.

Abstract

In this paper, we study the task of source-free domain adaptation (SFDA), where the source data are not available during target adaptation. Previous works on SFDA mainly focus on aligning the cross-domain distributions. However, they ignore the generalization ability of the pretrained source model, which largely influences the initial target outputs that are vital to the target adaptation stage. To address this, we make the interesting observation that the model accuracy is highly correlated with whether or not attention is focused on the objects in an image. To this end, we propose a generic and effective framework based on Transformer, named TransDA, for learning a generalized model for SFDA. Specifically, we apply the Transformer as the attention module and inject it into a convolutional network. By doing so, the model is encouraged to turn attention towards the object regions, which can effectively improve the model's generalization ability on the target domains. Moreover, a novel self-supervised knowledge distillation approach is proposed to adapt the Transformer with target pseudo-labels, thus further encouraging the network to focus on the object regions. Experiments on three domain adaptation tasks, including closed-set, partial-set, and open-set adaption, demonstrate that TransDA can greatly improve the adaptation accuracy and produce state-of-the-art results. The source code and trained models are available at https://github.com/ygjwd12345/TransDA.