Cross Aggregation Transformer for Image Restoration

TL;DR

This paper introduces the Cross Aggregation Transformer (CAT), a novel image restoration model that combines rectangle-window self-attention with local-global feature coupling to improve long-range dependency modeling and performance.

Contribution

The paper proposes the Cross Aggregation Transformer with Rectangle-Window Self-Attention and Axial-Shift, integrating CNN inductive biases for enhanced image restoration.

Findings

Outperforms state-of-the-art methods on multiple image restoration tasks

Effectively models long-range dependencies with rectangle-window attention

Enhances local-global feature integration through the Locality Complementary Module

Abstract

Recently, Transformer architecture has been introduced into image restoration to replace convolution neural network (CNN) with surprising results. Considering the high computational complexity of Transformer with global attention, some methods use the local square window to limit the scope of self-attention. However, these methods lack direct interaction among different windows, which limits the establishment of long-range dependencies. To address the above issue, we propose a new image restoration model, Cross Aggregation Transformer (CAT). The core of our CAT is the Rectangle-Window Self-Attention (Rwin-SA), which utilizes horizontal and vertical rectangle window attention in different heads parallelly to expand the attention area and aggregate the features cross different windows. We also introduce the Axial-Shift operation for different window interactions. Furthermore, we propose the Locality Complementary Module to complement the self-attention mechanism, which incorporates the inductive bias of CNN (e.g., translation invariance and locality) into Transformer, enabling global-local coupling. Extensive experiments demonstrate that our CAT outperforms recent state-of-the-art methods on several image restoration applications. The code and models are available at https://github.com/zhengchen1999/CAT.