UnSCAR: Universal, Scalable, Controllable, and Adaptable Image Restoration
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911496076787712 |
|---|---|
| author | Mandal, Debabrata Chattopadhyay, Soumitri Wang, Yujie Niethammer, Marc Chakravarthula, Praneeth |
| author_facet | Mandal, Debabrata Chattopadhyay, Soumitri Wang, Yujie Niethammer, Marc Chakravarthula, Praneeth |
| contents | Universal image restoration aims to recover clean images from arbitrary real-world degradations using a single inference model. Despite significant progress, existing all-in-one restoration networks do not scale to multiple degradations. As the number of degradations increases, training becomes unstable, models grow excessively large, and performance drops across both seen and unseen domains. In this work, we show that scaling universal restoration is fundamentally limited by interference across degradations during joint learning, leading to catastrophic task forgetting. To address this challenge, we introduce a unified inference pipeline with a multi-branch mixture-of-experts architecture that decomposes restoration knowledge across specialized task-adaptable experts. Our approach enables scalable learning (over sixteen degradations), adapts and generalizes robustly to unseen domains, and supports user-controllable restoration across degradations. Beyond achieving superior performance across benchmarks, this work establishes a new design paradigm for scalable and controllable universal image restoration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_07406 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UnSCAR: Universal, Scalable, Controllable, and Adaptable Image Restoration Mandal, Debabrata Chattopadhyay, Soumitri Wang, Yujie Niethammer, Marc Chakravarthula, Praneeth Computer Vision and Pattern Recognition Artificial Intelligence Universal image restoration aims to recover clean images from arbitrary real-world degradations using a single inference model. Despite significant progress, existing all-in-one restoration networks do not scale to multiple degradations. As the number of degradations increases, training becomes unstable, models grow excessively large, and performance drops across both seen and unseen domains. In this work, we show that scaling universal restoration is fundamentally limited by interference across degradations during joint learning, leading to catastrophic task forgetting. To address this challenge, we introduce a unified inference pipeline with a multi-branch mixture-of-experts architecture that decomposes restoration knowledge across specialized task-adaptable experts. Our approach enables scalable learning (over sixteen degradations), adapts and generalizes robustly to unseen domains, and supports user-controllable restoration across degradations. Beyond achieving superior performance across benchmarks, this work establishes a new design paradigm for scalable and controllable universal image restoration. |
| title | UnSCAR: Universal, Scalable, Controllable, and Adaptable Image Restoration |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2603.07406 |