Confidence-based Visual Dispersal for Few-shot Unsupervised Domain Adaptation

TL;DR

This paper introduces C-VisDiT, a confidence-based transfer learning method for few-shot unsupervised domain adaptation that selectively transfers high-confidence source knowledge and guides hard target samples with easier ones, improving adaptation performance.

Contribution

The paper proposes a novel confidence-based visual dispersal transfer learning approach (C-VisDiT) for FUDA, addressing the challenge of limited labeled source data and hard target samples.

Findings

C-VisDiT outperforms state-of-the-art FUDA methods on multiple benchmarks.

The method effectively transfers high-confidence source knowledge.

It improves classification of hard target samples.

Abstract

Unsupervised domain adaptation aims to transfer knowledge from a fully-labeled source domain to an unlabeled target domain. However, in real-world scenarios, providing abundant labeled data even in the source domain can be infeasible due to the difficulty and high expense of annotation. To address this issue, recent works consider the Few-shot Unsupervised Domain Adaptation (FUDA) where only a few source samples are labeled, and conduct knowledge transfer via self-supervised learning methods. Yet existing methods generally overlook that the sparse label setting hinders learning reliable source knowledge for transfer. Additionally, the learning difficulty difference in target samples is different but ignored, leaving hard target samples poorly classified. To tackle both deficiencies, in this paper, we propose a novel Confidence-based Visual Dispersal Transfer learning method (C-VisDiT) for FUDA. Specifically, C-VisDiT consists of a cross-domain visual dispersal strategy that transfers only high-confidence source knowledge for model adaptation and an intra-domain visual dispersal strategy that guides the learning of hard target samples with easy ones. We conduct extensive experiments on Office-31, Office-Home, VisDA-C, and DomainNet benchmark datasets and the results demonstrate that the proposed C-VisDiT significantly outperforms state-of-the-art FUDA methods. Our code is available at https://github.com/Bostoncake/C-VisDiT.