Image Restoration Learning via Noisy Supervision in the Fourier Domain

TL;DR

This paper introduces a Fourier domain-based learning framework for image restoration with noisy supervision, effectively handling spatially correlated noise and leveraging global Fourier coefficients for improved performance.

Contribution

It proposes a novel Fourier domain approach that models noise coefficients as Gaussian, enabling robust supervision and broad applicability across tasks and noise types.

Findings

Outperforms existing methods in quantitative metrics

Enhances perceptual image quality

Effectively handles spatially correlated noise

Abstract

Noisy supervision refers to supervising image restoration learning with noisy targets. It can alleviate the data collection burden and enhance the practical applicability of deep learning techniques. However, existing methods suffer from two key drawbacks. Firstly, they are ineffective in handling spatially correlated noise commonly observed in practical applications such as low-light imaging and remote sensing. Secondly, they rely on pixel-wise loss functions that only provide limited supervision information. This work addresses these challenges by leveraging the Fourier domain. We highlight that the Fourier coefficients of spatially correlated noise exhibit sparsity and independence, making them easier to handle. Additionally, Fourier coefficients contain global information, enabling more significant supervision. Motivated by these insights, we propose to establish noisy supervision in the Fourier domain. We first prove that Fourier coefficients of a wide range of noise converge in distribution to the Gaussian distribution. Exploiting this statistical property, we establish the equivalence between using noisy targets and clean targets in the Fourier domain. This leads to a unified learning framework applicable to various image restoration tasks, diverse network architectures, and different noise models. Extensive experiments validate the outstanding performance of this framework in terms of both quantitative indices and perceptual quality.