IntraStyler: Exemplar-based Style Synthesis for Cross-modality Domain Adaptation

2026-01-01

arXiv.org

AI Summary

This work addresses the limited style diversity in the target domain within cross-modal unsupervised domain adaptation by proposing an example-guided style synthesis method. Without requiring any prior knowledge, the approach dynamically guides the image translation process using exemplar images to generate diverse and controllable styles in the target domain. A contrastive learning–trained style encoder extracts pure style representations, which are integrated into an example-guided framework to effectively disentangle content and style. Experiments on the CrossMoDA 2023 dataset demonstrate that the proposed method significantly enhances the diversity of synthesized data and substantially improves performance on downstream segmentation tasks.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Multi-modal Vision
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Web Mining and Content Analysis: Large pretrained models with web data
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Image-level domain alignment is the de facto approach for unsupervised domain adaptation, where unpaired image translation is used to minimize the domain gap. Prior studies mainly focus on the domain shift between the source and target domains, whereas the intra-domain variability remains under-explored. To address the latter, an effective strategy is to diversify the styles of the synthetic target domain data during image translation. However, previous methods typically require intra-domain variations to be pre-specified for style synthesis, which may be impractical. In this paper, we propose an exemplar-based style synthesis method named IntraStyler, which can capture diverse intra-domain styles without any prior knowledge. Specifically, IntraStyler uses an exemplar image to guide the style synthesis such that the output style matches the exemplar style. To extract the style-only features, we introduce a style encoder to learn styles discriminatively based on contrastive learning. We evaluate the proposed method on the largest public dataset for cross-modality domain adaptation, CrossMoDA 2023. Our experiments show the efficacy of our method in controllable style synthesis and the benefits of diverse synthetic data for downstream segmentation. Code is available at https://github.com/han-liu/IntraStyler.

Problem

• intra-domain variability
• style synthesis
• cross-modality domain adaptation
• unsupervised domain adaptation
• image translation

Innovation

• exemplar-based style synthesis
• intra-domain variability
• contrastive learning
• unpaired image translation
• cross-modality domain adaptation