Deep Likelihood Network for Image Restoration with Multiple Degradation Levels

TL;DR

This paper introduces DL-Net, a deep likelihood network that enhances image restoration across multiple degradation levels by modifying existing networks with a recursive module, demonstrating improved performance in various tasks.

Contribution

The paper proposes a novel recursive module for existing networks to handle multiple degradation levels, improving generalization in image restoration tasks.

Findings

Effective in image inpainting, interpolation, and super-resolution

Outperforms single-degradation models across multiple levels

Demonstrates robustness and versatility in restoration tasks

Abstract

Convolutional neural networks have been proven effective in a variety of image restoration tasks. Most state-of-the-art solutions, however, are trained using images with a single particular degradation level, and their performance deteriorates drastically when applied to other degradation settings. In this paper, we propose deep likelihood network (DL-Net), aiming at generalizing off-the-shelf image restoration networks to succeed over a spectrum of degradation levels. We slightly modify an off-the-shelf network by appending a simple recursive module, which is derived from a fidelity term, for disentangling the computation for multiple degradation levels. Extensive experimental results on image inpainting, interpolation, and super-resolution show the effectiveness of our DL-Net.