UniVerse: A Unified Modulation Framework for Segmentation-Free,Disentangled Multi-Concept Personalization
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914619603288064 |
|---|---|
| author | Phung, Quynh Ghimire, Sandesh Hu, Minsi Tsai, Chung-Chi Huang, Jia-Bin |
| author_facet | Phung, Quynh Ghimire, Sandesh Hu, Minsi Tsai, Chung-Chi Huang, Jia-Bin |
| contents | Personalized visual understanding has advanced significantly, yet existing approaches struggle to localize and extract specific concepts when input images contain multiple objects. Many prior methods rely heavily on segmentation-based supervision or exhibit poor compositional generalization, limiting their ability to accurately disentangle and manipulate individual concepts. In this work, we propose UniVerse, a Unified Modulation Framework for segmentation-free, disentangled multi-concept personalization in diffusion transformers. Our method allows for composable and decomposable concept extraction, enabling fine-grained localization and representation of target objects without explicit segmentation masks. UniVerse learns to decompose complex scenes into concept-specific representations and then compose them in a unified manner, enabling robust personalization across diverse visual contexts. Through extensive experiments on multiple benchmarks, we demonstrate that UniVerse significantly outperforms state-of-the-art baselines in both localization accuracy and visual fidelity. Qualitative and quantitative results show that our approach can precisely extract target concepts in cluttered scenes, paving the way for more flexible, interpretable, and personalized visual generation and understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2606_00351 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UniVerse: A Unified Modulation Framework for Segmentation-Free,Disentangled Multi-Concept Personalization Phung, Quynh Ghimire, Sandesh Hu, Minsi Tsai, Chung-Chi Huang, Jia-Bin Computer Vision and Pattern Recognition Personalized visual understanding has advanced significantly, yet existing approaches struggle to localize and extract specific concepts when input images contain multiple objects. Many prior methods rely heavily on segmentation-based supervision or exhibit poor compositional generalization, limiting their ability to accurately disentangle and manipulate individual concepts. In this work, we propose UniVerse, a Unified Modulation Framework for segmentation-free, disentangled multi-concept personalization in diffusion transformers. Our method allows for composable and decomposable concept extraction, enabling fine-grained localization and representation of target objects without explicit segmentation masks. UniVerse learns to decompose complex scenes into concept-specific representations and then compose them in a unified manner, enabling robust personalization across diverse visual contexts. Through extensive experiments on multiple benchmarks, we demonstrate that UniVerse significantly outperforms state-of-the-art baselines in both localization accuracy and visual fidelity. Qualitative and quantitative results show that our approach can precisely extract target concepts in cluttered scenes, paving the way for more flexible, interpretable, and personalized visual generation and understanding. |
| title | UniVerse: A Unified Modulation Framework for Segmentation-Free,Disentangled Multi-Concept Personalization |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2606.00351 |