StyleDecoupler: Generalizable Artistic Style Disentanglement
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908786371854336 |
|---|---|
| author | Jia, Zexi Zhang, Jinchao Zhou, Jie |
| author_facet | Jia, Zexi Zhang, Jinchao Zhou, Jie |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_17697 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | StyleDecoupler: Generalizable Artistic Style Disentanglement Jia, Zexi Zhang, Jinchao Zhou, Jie Computer Vision and Pattern Recognition 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. |
| title | StyleDecoupler: Generalizable Artistic Style Disentanglement |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.17697 |