DAFD: Domain Adaptation via Feature Disentanglement for Image Classification

TL;DR

This paper introduces DAFD, a novel method for unsupervised domain adaptation in image classification that disentangles category-relevant features and aligns them across domains, improving accuracy under domain shift.

Contribution

The paper proposes a new feature disentanglement approach with CRFS and DLMMD modules for better domain alignment in unsupervised image classification.

Findings

Outperforms existing methods on four standard datasets.

Effectively disentangles relevant features to reduce domain discrepancy.

Improves classification accuracy in domain adaptation scenarios.

Abstract

A good feature representation is the key to image classification. In practice, image classifiers may be applied in scenarios different from what they have been trained on. This so-called domain shift leads to a significant performance drop in image classification. Unsupervised domain adaptation (UDA) reduces the domain shift by transferring the knowledge learned from a labeled source domain to an unlabeled target domain. We perform feature disentanglement for UDA by distilling category-relevant features and excluding category-irrelevant features from the global feature maps. This disentanglement prevents the network from overfitting to category-irrelevant information and makes it focus on information useful for classification. This reduces the difficulty of domain alignment and improves the classification accuracy on the target domain. We propose a coarse-to-fine domain adaptation method called Domain Adaptation via Feature Disentanglement~(DAFD), which has two components: (1)the Category-Relevant Feature Selection (CRFS) module, which disentangles the category-relevant features from the category-irrelevant features, and (2)the Dynamic Local Maximum Mean Discrepancy (DLMMD) module, which achieves fine-grained alignment by reducing the discrepancy within the category-relevant features from different domains. Combined with the CRFS, the DLMMD module can align the category-relevant features properly. We conduct comprehensive experiment on four standard datasets. Our results clearly demonstrate the robustness and effectiveness of our approach in domain adaptive image classification tasks and its competitiveness to the state of the art.