GRFormer: Grouped Residual Self-Attention for Lightweight Single Image Super-Resolution

TL;DR

GRFormer introduces a lightweight, efficient self-attention mechanism with grouped residual layers and a novel position bias, significantly improving super-resolution performance while reducing parameters and computations.

Contribution

The paper proposes GRFormer, a novel self-attention module with grouped residual layers and a new position bias, enhancing efficiency and performance in image super-resolution.

Findings

Outperforms state-of-the-art methods in PSNR on DIV2K dataset.

Reduces parameters and MACs by approximately 60% and 49%.

Achieves up to 0.23dB PSNR improvement.

Abstract

Previous works have shown that reducing parameter overhead and computations for transformer-based single image super-resolution (SISR) models (e.g., SwinIR) usually leads to a reduction of performance. In this paper, we present GRFormer, an efficient and lightweight method, which not only reduces the parameter overhead and computations, but also greatly improves performance. The core of GRFormer is Grouped Residual Self-Attention (GRSA), which is specifically oriented towards two fundamental components. Firstly, it introduces a novel grouped residual layer (GRL) to replace the Query, Key, Value (QKV) linear layer in self-attention, aimed at efficiently reducing parameter overhead, computations, and performance loss at the same time. Secondly, it integrates a compact Exponential-Space Relative Position Bias (ES-RPB) as a substitute for the original relative position bias to improve the ability to represent position information while further minimizing the parameter count. Extensive experimental results demonstrate that GRFormer outperforms state-of-the-art transformer-based methods for $\times$2, $\times$3 and $\times$4 SISR tasks, notably outperforming SOTA by a maximum PSNR of 0.23dB when trained on the DIV2K dataset, while reducing the number of parameter and MACs by about \textbf{60\%} and \textbf{49\% } in only self-attention module respectively. We hope that our simple and effective method that can easily applied to SR models based on window-division self-attention can serve as a useful tool for further research in image super-resolution. The code is available at \url{https://github.com/sisrformer/GRFormer}.