Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911825285611520 |
|---|---|
| author | Yu, Fanghua Gu, Jinjin Li, Zheyuan Hu, Jinfan Kong, Xiangtao Wang, Xintao He, Jingwen Qiao, Yu Dong, Chao |
| author_facet | Yu, Fanghua Gu, Jinjin Li, Zheyuan Hu, Jinfan Kong, Xiangtao Wang, Xintao He, Jingwen Qiao, Yu Dong, Chao |
| contents | We introduce SUPIR (Scaling-UP Image Restoration), a groundbreaking image restoration method that harnesses generative prior and the power of model scaling up. Leveraging multi-modal techniques and advanced generative prior, SUPIR marks a significant advance in intelligent and realistic image restoration. As a pivotal catalyst within SUPIR, model scaling dramatically enhances its capabilities and demonstrates new potential for image restoration. We collect a dataset comprising 20 million high-resolution, high-quality images for model training, each enriched with descriptive text annotations. SUPIR provides the capability to restore images guided by textual prompts, broadening its application scope and potential. Moreover, we introduce negative-quality prompts to further improve perceptual quality. We also develop a restoration-guided sampling method to suppress the fidelity issue encountered in generative-based restoration. Experiments demonstrate SUPIR's exceptional restoration effects and its novel capacity to manipulate restoration through textual prompts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_13627 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild Yu, Fanghua Gu, Jinjin Li, Zheyuan Hu, Jinfan Kong, Xiangtao Wang, Xintao He, Jingwen Qiao, Yu Dong, Chao Computer Vision and Pattern Recognition We introduce SUPIR (Scaling-UP Image Restoration), a groundbreaking image restoration method that harnesses generative prior and the power of model scaling up. Leveraging multi-modal techniques and advanced generative prior, SUPIR marks a significant advance in intelligent and realistic image restoration. As a pivotal catalyst within SUPIR, model scaling dramatically enhances its capabilities and demonstrates new potential for image restoration. We collect a dataset comprising 20 million high-resolution, high-quality images for model training, each enriched with descriptive text annotations. SUPIR provides the capability to restore images guided by textual prompts, broadening its application scope and potential. Moreover, we introduce negative-quality prompts to further improve perceptual quality. We also develop a restoration-guided sampling method to suppress the fidelity issue encountered in generative-based restoration. Experiments demonstrate SUPIR's exceptional restoration effects and its novel capacity to manipulate restoration through textual prompts. |
| title | Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2401.13627 |