Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2504.00996 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908294368460800 |
|---|---|
| author | Xie, Liangbin Pakhomov, Daniil Wang, Zhonghao Wu, Zongze Chen, Ziyan Zhou, Yuqian Zheng, Haitian Zhang, Zhifei Lin, Zhe Zhou, Jiantao Dong, Chao |
| author_facet | Xie, Liangbin Pakhomov, Daniil Wang, Zhonghao Wu, Zongze Chen, Ziyan Zhou, Yuqian Zheng, Haitian Zhang, Zhifei Lin, Zhe Zhou, Jiantao Dong, Chao |
| contents | This paper introduces TurboFill, a fast image inpainting model that enhances a few-step text-to-image diffusion model with an inpainting adapter for high-quality and efficient inpainting. While standard diffusion models generate high-quality results, they incur high computational costs. We overcome this by training an inpainting adapter on a few-step distilled text-to-image model, DMD2, using a novel 3-step adversarial training scheme to ensure realistic, structurally consistent, and visually harmonious inpainted regions. To evaluate TurboFill, we propose two benchmarks: DilationBench, which tests performance across mask sizes, and HumanBench, based on human feedback for complex prompts. Experiments show that TurboFill outperforms both multi-step BrushNet and few-step inpainting methods, setting a new benchmark for high-performance inpainting tasks. Our project page: https://liangbinxie.github.io/projects/TurboFill/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_00996 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | TurboFill: Adapting Few-step Text-to-image Model for Fast Image Inpainting Xie, Liangbin Pakhomov, Daniil Wang, Zhonghao Wu, Zongze Chen, Ziyan Zhou, Yuqian Zheng, Haitian Zhang, Zhifei Lin, Zhe Zhou, Jiantao Dong, Chao Computer Vision and Pattern Recognition This paper introduces TurboFill, a fast image inpainting model that enhances a few-step text-to-image diffusion model with an inpainting adapter for high-quality and efficient inpainting. While standard diffusion models generate high-quality results, they incur high computational costs. We overcome this by training an inpainting adapter on a few-step distilled text-to-image model, DMD2, using a novel 3-step adversarial training scheme to ensure realistic, structurally consistent, and visually harmonious inpainted regions. To evaluate TurboFill, we propose two benchmarks: DilationBench, which tests performance across mask sizes, and HumanBench, based on human feedback for complex prompts. Experiments show that TurboFill outperforms both multi-step BrushNet and few-step inpainting methods, setting a new benchmark for high-performance inpainting tasks. Our project page: https://liangbinxie.github.io/projects/TurboFill/ |
| title | TurboFill: Adapting Few-step Text-to-image Model for Fast Image Inpainting |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.00996 |