SUMD: Super U-shaped Matrix Decomposition Convolutional neural network for Image denoising

TL;DR

SUMD introduces a novel CNN framework with a matrix decomposition module and U-shaped architecture to effectively capture local and global features, achieving Transformer-level performance in image denoising.

Contribution

The paper presents a new CNN-based image denoising method that incorporates a matrix decomposition module within a U-shaped architecture to model global context.

Findings

Achieves comparable results to Transformer-based methods on multiple datasets.

Effectively captures global features using matrix decomposition within CNN.

Improves denoising quality with a multi-stage progressive restoration approach.

Abstract

In this paper, we propose a novel and efficient CNN-based framework that leverages local and global context information for image denoising. Due to the limitations of convolution itself, the CNN-based method is generally unable to construct an effective and structured global feature representation, usually called the long-distance dependencies in the Transformer-based method. To tackle this problem, we introduce the matrix decomposition module(MD) in the network to establish the global context feature, comparable to the Transformer based method performance. Inspired by the design of multi-stage progressive restoration of U-shaped architecture, we further integrate the MD module into the multi-branches to acquire the relative global feature representation of the patch range at the current stage. Then, the stage input gradually rises to the overall scope and continuously improves the final feature. Experimental results on various image denoising datasets: SIDD, DND, and synthetic Gaussian noise datasets show that our model(SUMD) can produce comparable visual quality and accuracy results with Transformer-based methods.