Multi-level Attention-guided Graph Neural Network for Image Restoration

TL;DR

This paper introduces a multi-level attention-guided graph neural network that explicitly models local and global features for improved image restoration, achieving state-of-the-art results.

Contribution

It proposes a novel graph neural network architecture that explicitly constructs local and global feature graphs with multi-attention mechanisms for image restoration.

Findings

Achieves state-of-the-art performance on image restoration tasks.

Effectively models local and global features through graph-based attention mechanisms.

Improves restoration quality by integrating local structure and global information.

Abstract

In recent years, deep learning has achieved remarkable success in the field of image restoration. However, most convolutional neural network-based methods typically focus on a single scale, neglecting the incorporation of multi-scale information. In image restoration tasks, local features of an image are often insufficient, necessitating the integration of global features to complement them. Although recent neural network algorithms have made significant strides in feature extraction, many models do not explicitly model global features or consider the relationship between global and local features. This paper proposes multi-level attention-guided graph neural network. The proposed network explicitly constructs element block graphs and element graphs within feature maps using multi-attention mechanisms to extract both local structural features and global representation information of the image. Since the network struggles to effectively extract global information during image degradation, the structural information of local feature blocks can be used to correct and supplement the global information. Similarly, when element block information in the feature map is missing, it can be refined using global element representation information. The graph within the network learns real-time dynamic connections through the multi-attention mechanism, and information is propagated and aggregated via graph convolution algorithms. By combining local element block information and global element representation information from the feature map, the algorithm can more effectively restore missing information in the image. Experimental results on several classic image restoration tasks demonstrate the effectiveness of the proposed method, achieving state-of-the-art performance.