Efficient and Explicit Modelling of Image Hierarchies for Image Restoration
By Yawei Li, Yuchen Fan, Xiaoyu Xiang, Denis Demandolx, Rakesh Ranjan, Radu Timofte, Luc Van Gool. Computer Vision and Pattern Recognition, 2023
340 citations recorded by the source corpus.
The aim of this paper is to propose a mechanism to efficiently and explicitly model image hierarchies in the global, regional, and local range for image restoration. To achieve that, we start by analyzing two important properties of natural images including cross-scale similarity and anisotropic image features. Inspired by that, we propose the anchored stripe self-attention which achieves a good balance between the space and time complexity of self-attention and the modelling capacity beyond the regional range. Then we propose a new network architecture dubbed GRL to explicitly model image hierarchies in the Global, Regional, and Local range via anchored stripe self-attention, window self-attention, and channel attention enhanced convolution. Finally, the proposed network is applied to 7 image restoration types, covering both real and synthetic settings. The proposed method sets the new state-of-the-art for several of those. Code will be available at https://github.com/ofsoundof/GRL-Image-Restoration.git.
Subfields: Computer Vision and Pattern Recognition, Media Technology
View Paper: https://www.semanticscholar.org/paper/2920b885278a6973a306b57201e3fc3273b71132
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computer Vision and Pattern Recognition
In Media Technology
Citations in the corpus
- MambaIR: A Simple Baseline for Image Restoration with State-Space Model (2024, methodology): “With the introduction of modern deep learning models such as CNNs [13,16,42,81,89] and Transformers [8,10,12,40,41], state-of-the-art performance has continued to be refreshed in the past few years.”
- LocalViT: Analyzing Locality in Vision Transformers (2021, neutral): “A parallel research strand incorporates global connectivity into the network via self-attention [4], [7], [8], [9], [10].”
- Omni-Kernel Network for Image Restoration (2024, methodology): “In addition to the regular depth-wise convolution, inspired by strip-based self-attention (Dong et al. 2022; Tsai et al. 2022; Li et al. 2023), we also employ 1 × K and K × 1 depth-wise convolutions in parallel to the square one to harvest strip-shaped contextual information.”