Image Super-Resolution via RL-CSC: When Residual Learning Meets Convolutional Sparse Coding

TL;DR

This paper introduces RL-CSC, a convolutional sparse coding-based model for single image super-resolution that combines residual learning to improve interpretability and training, achieving state-of-the-art results.

Contribution

The paper extends LISTA to a convolutional version, integrating residual learning into a deep sparse coding network for enhanced super-resolution performance.

Findings

RL-CSC outperforms recent state-of-the-art methods in accuracy.

The model effectively recovers high-frequency details.

Residual learning facilitates training deeper networks.

Abstract

We propose a simple yet effective model for Single Image Super-Resolution (SISR), by combining the merits of Residual Learning and Convolutional Sparse Coding (RL-CSC). Our model is inspired by the Learned Iterative Shrinkage-Threshold Algorithm (LISTA). We extend LISTA to its convolutional version and build the main part of our model by strictly following the convolutional form, which improves the network's interpretability. Specifically, the convolutional sparse codings of input feature maps are learned in a recursive manner, and high-frequency information can be recovered from these CSCs. More importantly, residual learning is applied to alleviate the training difficulty when the network goes deeper. Extensive experiments on benchmark datasets demonstrate the effectiveness of our method. RL-CSC (30 layers) outperforms several recent state-of-the-arts, e.g., DRRN (52 layers) and MemNet (80 layers) in both accuracy and visual qualities. Codes and more results are available at https://github.com/axzml/RL-CSC.