Guided Frequency Loss for Image Restoration

TL;DR

This paper introduces the Guided Frequency Loss (GFL), a novel loss function that enhances image restoration models by balancing frequency and spatial content learning, leading to improved PSNR in super-resolution and denoising tasks.

Contribution

The paper proposes the GFL, combining three components to improve frequency domain learning in image restoration models, which was underexplored in prior work.

Findings

GFL improved PSNR across multiple datasets and architectures.

GFL enhanced training efficiency for super-resolution models.

GFL was particularly effective on constrained data with less stochasticity.

Abstract

Image Restoration has seen remarkable progress in recent years. Many generative models have been adapted to tackle the known restoration cases of images. However, the interest in benefiting from the frequency domain is not well explored despite its major factor in these particular cases of image synthesis. In this study, we propose the Guided Frequency Loss (GFL), which helps the model to learn in a balanced way the image's frequency content alongside the spatial content. It aggregates three major components that work in parallel to enhance learning efficiency; a Charbonnier component, a Laplacian Pyramid component, and a Gradual Frequency component. We tested GFL on the Super Resolution and the Denoising tasks. We used three different datasets and three different architectures for each of them. We found that the GFL loss improved the PSNR metric in most implemented experiments. Also, it improved the training of the Super Resolution models in both SwinIR and SRGAN. In addition, the utility of the GFL loss increased better on constrained data due to the less stochasticity in the high frequencies' components among samples.