Unified-Width Adaptive Dynamic Network for All-In-One Image Restoration
Yimin Xu, Nanxi Gao, Zhongyun Shan, Fei Chao, Rongrong Ji

TL;DR
The paper introduces U-WADN, a dynamic network that adapts its width for all-in-one image restoration tasks, improving performance and efficiency by automatically selecting optimal sub-network widths based on task complexity.
Contribution
It proposes a novel unified adaptive network with width selection for diverse image restoration tasks, addressing task complexity and resource allocation issues.
Findings
Achieves better restoration performance across tasks.
Reduces up to 32.3% FLOPs during inference.
Provides approximately 15.7% real-time acceleration.
Abstract
In contrast to traditional image restoration methods, all-in-one image restoration techniques are gaining increased attention for their ability to restore images affected by diverse and unknown corruption types and levels. However, contemporary all-in-one image restoration methods omit task-wise difficulties and employ the same networks to reconstruct images afflicted by diverse degradations. This practice leads to an underestimation of the task correlations and suboptimal allocation of computational resources. To elucidate task-wise complexities, we introduce a novel concept positing that intricate image degradation can be represented in terms of elementary degradation. Building upon this foundation, we propose an innovative approach, termed the Unified-Width Adaptive Dynamic Network (U-WADN), consisting of two pivotal components: a Width Adaptive Backbone (WAB) and a Width Selector…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdvanced Image Processing Techniques · Advanced Vision and Imaging · Image Enhancement Techniques
