On Fine-Tuned Deep Features for Unsupervised Domain Adaptation

TL;DR

This paper investigates combining fine-tuned deep features with feature transformation methods for unsupervised domain adaptation, demonstrating improved performance over existing approaches through extensive experiments.

Contribution

It introduces a novel approach that integrates fine-tuning with feature transformation based UDA, achieving state-of-the-art results on benchmark datasets.

Findings

Fine-tuned features enhance domain adaptation performance.

Combining fine-tuning with SPL yields superior results.

Experiments on multiple models and datasets confirm effectiveness.

Abstract

Prior feature transformation based approaches to Unsupervised Domain Adaptation (UDA) employ the deep features extracted by pre-trained deep models without fine-tuning them on the specific source or target domain data for a particular domain adaptation task. In contrast, end-to-end learning based approaches optimise the pre-trained backbones and the customised adaptation modules simultaneously to learn domain-invariant features for UDA. In this work, we explore the potential of combining fine-tuned features and feature transformation based UDA methods for improved domain adaptation performance. Specifically, we integrate the prevalent progressive pseudo-labelling techniques into the fine-tuning framework to extract fine-tuned features which are subsequently used in a state-of-the-art feature transformation based domain adaptation method SPL (Selective Pseudo-Labeling). Thorough experiments with multiple deep models including ResNet-50/101 and DeiT-small/base are conducted to demonstrate the combination of fine-tuned features and SPL can achieve state-of-the-art performance on several benchmark datasets.