Cat-AIR: Content and Task-Aware All-in-One Image Restoration

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Jiang, Jiachen, Ding, Tianyu, Zhang, Ke, Zhou, Jinxin, Chen, Tianyi, Zharkov, Ilya, Zhu, Zhihui, Liang, Luming
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916660377550848
author Jiang, Jiachen
Ding, Tianyu
Zhang, Ke
Zhou, Jinxin
Chen, Tianyi
Zharkov, Ilya
Zhu, Zhihui
Liang, Luming
author_facet Jiang, Jiachen
Ding, Tianyu
Zhang, Ke
Zhou, Jinxin
Chen, Tianyi
Zharkov, Ilya
Zhu, Zhihui
Liang, Luming
contents All-in-one image restoration seeks to recover high-quality images from various types of degradation using a single model, without prior knowledge of the corruption source. However, existing methods often struggle to effectively and efficiently handle multiple degradation types. We present Cat-AIR, a novel \textbf{C}ontent \textbf{A}nd \textbf{T}ask-aware framework for \textbf{A}ll-in-one \textbf{I}mage \textbf{R}estoration. Cat-AIR incorporates an alternating spatial-channel attention mechanism that adaptively balances the local and global information for different tasks. Specifically, we introduce cross-layer channel attentions and cross-feature spatial attentions that allocate computations based on content and task complexity. Furthermore, we propose a smooth learning strategy that allows for seamless adaptation to new restoration tasks while maintaining performance on existing ones. Extensive experiments demonstrate that Cat-AIR achieves state-of-the-art results across a wide range of restoration tasks, requiring fewer FLOPs than previous methods, establishing new benchmarks for efficient all-in-one image restoration.
format Preprint
id arxiv_https___arxiv_org_abs_2503_17915
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cat-AIR: Content and Task-Aware All-in-One Image Restoration
Jiang, Jiachen
Ding, Tianyu
Zhang, Ke
Zhou, Jinxin
Chen, Tianyi
Zharkov, Ilya
Zhu, Zhihui
Liang, Luming
Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
All-in-one image restoration seeks to recover high-quality images from various types of degradation using a single model, without prior knowledge of the corruption source. However, existing methods often struggle to effectively and efficiently handle multiple degradation types. We present Cat-AIR, a novel \textbf{C}ontent \textbf{A}nd \textbf{T}ask-aware framework for \textbf{A}ll-in-one \textbf{I}mage \textbf{R}estoration. Cat-AIR incorporates an alternating spatial-channel attention mechanism that adaptively balances the local and global information for different tasks. Specifically, we introduce cross-layer channel attentions and cross-feature spatial attentions that allocate computations based on content and task complexity. Furthermore, we propose a smooth learning strategy that allows for seamless adaptation to new restoration tasks while maintaining performance on existing ones. Extensive experiments demonstrate that Cat-AIR achieves state-of-the-art results across a wide range of restoration tasks, requiring fewer FLOPs than previous methods, establishing new benchmarks for efficient all-in-one image restoration.
title Cat-AIR: Content and Task-Aware All-in-One Image Restoration
topic Image and Video Processing
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2503.17915