Semi-Supervised Domain Adaptation via Selective Pseudo Labeling and Progressive Self-Training

TL;DR

This paper introduces a semi-supervised domain adaptation method that leverages labeled target images for selective pseudo labeling and employs a noise-robust learning scheme to improve adaptation performance.

Contribution

It proposes a novel SSDA approach that uses labeled target data for selective pseudo labeling and progressive self-training with noise robustness, outperforming previous methods.

Findings

Outperforms previous SSDA methods in experiments.

Effectively utilizes labeled target images for pseudo label generation.

Progressively updates model and pseudo labels to handle noise.

Abstract

Domain adaptation (DA) is a representation learning methodology that transfers knowledge from a label-sufficient source domain to a label-scarce target domain. While most of early methods are focused on unsupervised DA (UDA), several studies on semi-supervised DA (SSDA) are recently suggested. In SSDA, a small number of labeled target images are given for training, and the effectiveness of those data is demonstrated by the previous studies. However, the previous SSDA approaches solely adopt those data for embedding ordinary supervised losses, overlooking the potential usefulness of the few yet informative clues. Based on this observation, in this paper, we propose a novel method that further exploits the labeled target images for SSDA. Specifically, we utilize labeled target images to selectively generate pseudo labels for unlabeled target images. In addition, based on the observation that pseudo labels are inevitably noisy, we apply a label noise-robust learning scheme, which progressively updates the network and the set of pseudo labels by turns. Extensive experimental results show that our proposed method outperforms other previous state-of-the-art SSDA methods.