CRNet: Image Super-Resolution Using A Convolutional Sparse Coding Inspired Network

TL;DR

This paper introduces CRNet, a novel convolutional neural network inspired by convolutional sparse coding, to improve image super-resolution by leveraging global image correlations, demonstrating superior results over existing methods.

Contribution

The paper develops a CSC-inspired CNN framework for image super-resolution, integrating the Convolutional Iterative Soft Thresholding Algorithm into neural network architecture.

Findings

Outperforms state-of-the-art SR methods in quantitative metrics.

Effective in both pre- and post-upsampling scenarios.

Shows superior qualitative image reconstruction quality.

Abstract

Convolutional Sparse Coding (CSC) has been attracting more and more attention in recent years, for making full use of image global correlation to improve performance on various computer vision applications. However, very few studies focus on solving CSC based image Super-Resolution (SR) problem. As a consequence, there is no significant progress in this area over a period of time. In this paper, we exploit the natural connection between CSC and Convolutional Neural Networks (CNN) to address CSC based image SR. Specifically, Convolutional Iterative Soft Thresholding Algorithm (CISTA) is introduced to solve CSC problem and it can be implemented using CNN architectures. Then we develop a novel CSC based SR framework analogy to the traditional SC based SR methods. Two models inspired by this framework are proposed for pre-/post-upsampling SR, respectively. Compared with recent state-of-the-art SR methods, both of our proposed models show superior performance in terms of both quantitative and qualitative measurements.