Source Free Domain Adaptation with Image Translation

TL;DR

This paper introduces a source-free domain adaptation method that uses image translation to align target image styles with source styles, improving classification accuracy without access to source data.

Contribution

It proposes a novel style transfer approach for source-free domain adaptation that aligns feature statistics to adapt pre-trained models to new target domains.

Findings

Consistent and statistically significant accuracy improvements on multiple datasets.

Effective style transfer enhances pre-trained model performance in target domains.

Method outperforms existing source-free domain adaptation techniques.

Abstract

Effort in releasing large-scale datasets may be compromised by privacy and intellectual property considerations. A feasible alternative is to release pre-trained models instead. While these models are strong on their original task (source domain), their performance might degrade significantly when deployed directly in a new environment (target domain), which might not contain labels for training under realistic settings. Domain adaptation (DA) is a known solution to the domain gap problem, but usually requires labeled source data. In this paper, we study the problem of source free domain adaptation (SFDA), whose distinctive feature is that the source domain only provides a pre-trained model, but no source data. Being source free adds significant challenges to DA, especially when considering that the target dataset is unlabeled. To solve the SFDA problem, we propose an image translation approach that transfers the style of target images to that of unseen source images. To this end, we align the batch-wise feature statistics of generated images to that stored in batch normalization layers of the pre-trained model. Compared with directly classifying target images, higher accuracy is obtained with these style transferred images using the pre-trained model. On several image classification datasets, we show that the above-mentioned improvements are consistent and statistically significant.