RealCompo: Balancing Realism and Compositionality Improves Text-to-Image Diffusion Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909347155542016 |
|---|---|
| author | Zhang, Xinchen Yang, Ling Cai, Yaqi Yu, Zhaochen Wang, Kai-Ni Xie, Jiake Tian, Ye Xu, Minkai Tang, Yong Yang, Yujiu Cui, Bin |
| author_facet | Zhang, Xinchen Yang, Ling Cai, Yaqi Yu, Zhaochen Wang, Kai-Ni Xie, Jiake Tian, Ye Xu, Minkai Tang, Yong Yang, Yujiu Cui, Bin |
| contents | Diffusion models have achieved remarkable advancements in text-to-image generation. However, existing models still have many difficulties when faced with multiple-object compositional generation. In this paper, we propose RealCompo, a new training-free and transferred-friendly text-to-image generation framework, which aims to leverage the respective advantages of text-to-image models and spatial-aware image diffusion models (e.g., layout, keypoints and segmentation maps) to enhance both realism and compositionality of the generated images. An intuitive and novel balancer is proposed to dynamically balance the strengths of the two models in denoising process, allowing plug-and-play use of any model without extra training. Extensive experiments show that our RealCompo consistently outperforms state-of-the-art text-to-image models and spatial-aware image diffusion models in multiple-object compositional generation while keeping satisfactory realism and compositionality of the generated images. Notably, our RealCompo can be seamlessly extended with a wide range of spatial-aware image diffusion models and stylized diffusion models. Our code is available at: https://github.com/YangLing0818/RealCompo |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_12908 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | RealCompo: Balancing Realism and Compositionality Improves Text-to-Image Diffusion Models Zhang, Xinchen Yang, Ling Cai, Yaqi Yu, Zhaochen Wang, Kai-Ni Xie, Jiake Tian, Ye Xu, Minkai Tang, Yong Yang, Yujiu Cui, Bin Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Diffusion models have achieved remarkable advancements in text-to-image generation. However, existing models still have many difficulties when faced with multiple-object compositional generation. In this paper, we propose RealCompo, a new training-free and transferred-friendly text-to-image generation framework, which aims to leverage the respective advantages of text-to-image models and spatial-aware image diffusion models (e.g., layout, keypoints and segmentation maps) to enhance both realism and compositionality of the generated images. An intuitive and novel balancer is proposed to dynamically balance the strengths of the two models in denoising process, allowing plug-and-play use of any model without extra training. Extensive experiments show that our RealCompo consistently outperforms state-of-the-art text-to-image models and spatial-aware image diffusion models in multiple-object compositional generation while keeping satisfactory realism and compositionality of the generated images. Notably, our RealCompo can be seamlessly extended with a wide range of spatial-aware image diffusion models and stylized diffusion models. Our code is available at: https://github.com/YangLing0818/RealCompo |
| title | RealCompo: Balancing Realism and Compositionality Improves Text-to-Image Diffusion Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2402.12908 |