UniCombine: Unified Multi-Conditional Combination with Diffusion Transformer
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912470583476224 |
|---|---|
| author | Wang, Haoxuan Peng, Jinlong He, Qingdong Yang, Hao Jin, Ying Wu, Jiafu Hu, Xiaobin Pan, Yanjie Gan, Zhenye Chi, Mingmin Peng, Bo Wang, Yabiao |
| author_facet | Wang, Haoxuan Peng, Jinlong He, Qingdong Yang, Hao Jin, Ying Wu, Jiafu Hu, Xiaobin Pan, Yanjie Gan, Zhenye Chi, Mingmin Peng, Bo Wang, Yabiao |
| contents | With the rapid development of diffusion models in image generation, the demand for more powerful and flexible controllable frameworks is increasing. Although existing methods can guide generation beyond text prompts, the challenge of effectively combining multiple conditional inputs while maintaining consistency with all of them remains unsolved. To address this, we introduce UniCombine, a DiT-based multi-conditional controllable generative framework capable of handling any combination of conditions, including but not limited to text prompts, spatial maps, and subject images. Specifically, we introduce a novel Conditional MMDiT Attention mechanism and incorporate a trainable LoRA module to build both the training-free and training-based versions. Additionally, we propose a new pipeline to construct SubjectSpatial200K, the first dataset designed for multi-conditional generative tasks covering both the subject-driven and spatially-aligned conditions. Extensive experimental results on multi-conditional generation demonstrate the outstanding universality and powerful capability of our approach with state-of-the-art performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_09277 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | UniCombine: Unified Multi-Conditional Combination with Diffusion Transformer Wang, Haoxuan Peng, Jinlong He, Qingdong Yang, Hao Jin, Ying Wu, Jiafu Hu, Xiaobin Pan, Yanjie Gan, Zhenye Chi, Mingmin Peng, Bo Wang, Yabiao Computer Vision and Pattern Recognition Artificial Intelligence With the rapid development of diffusion models in image generation, the demand for more powerful and flexible controllable frameworks is increasing. Although existing methods can guide generation beyond text prompts, the challenge of effectively combining multiple conditional inputs while maintaining consistency with all of them remains unsolved. To address this, we introduce UniCombine, a DiT-based multi-conditional controllable generative framework capable of handling any combination of conditions, including but not limited to text prompts, spatial maps, and subject images. Specifically, we introduce a novel Conditional MMDiT Attention mechanism and incorporate a trainable LoRA module to build both the training-free and training-based versions. Additionally, we propose a new pipeline to construct SubjectSpatial200K, the first dataset designed for multi-conditional generative tasks covering both the subject-driven and spatially-aligned conditions. Extensive experimental results on multi-conditional generation demonstrate the outstanding universality and powerful capability of our approach with state-of-the-art performance. |
| title | UniCombine: Unified Multi-Conditional Combination with Diffusion Transformer |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2503.09277 |