StyleDecoupler: Generalizable Artistic Style Disentanglement

TL;DR

StyleDecoupler is a novel framework that disentangles artistic style from content using uni-modal and multi-modal vision models, enabling improved style retrieval and analysis without model fine-tuning.

Contribution

It introduces a universal, plug-and-play style disentanglement method leveraging information theory and uni-modal representations, along with a large-scale artwork dataset for benchmarking.

Findings

Achieves state-of-the-art style retrieval performance

Enables style relationship mapping and generative model evaluation

Operates without fine-tuning on frozen vision-language models

Abstract

Representing artistic style is challenging due to its deep entanglement with semantic content. We propose StyleDecoupler, an information-theoretic framework that leverages a key insight: multi-modal vision models encode both style and content, while uni-modal models suppress style to focus on content-invariant features. By using uni-modal representations as content-only references, we isolate pure style features from multi-modal embeddings through mutual information minimization. StyleDecoupler operates as a plug-and-play module on frozen Vision-Language Models without fine-tuning. We also introduce WeART, a large-scale benchmark of 280K artworks across 152 styles and 1,556 artists. Experiments show state-of-the-art performance on style retrieval across WeART and WikiART, while enabling applications like style relationship mapping and generative model evaluation. We release our method and dataset at this url.