Disentangling Noise from Images: A Flow-Based Image Denoising Neural Network

TL;DR

This paper introduces an invertible flow-based neural network for image denoising that models and disentangles noise and clean image distributions, outperforming traditional CNN methods in accuracy, speed, and parameter efficiency.

Contribution

It proposes a novel distribution learning framework and an invertible neural network, FDN, for image denoising without assumptions on noise or clean image distributions.

Findings

FDN effectively removes synthetic Gaussian noise from images.

FDN surpasses previous methods in real image denoising performance.

FDN achieves faster processing with fewer parameters.

Abstract

The prevalent convolutional neural network (CNN) based image denoising methods extract features of images to restore the clean ground truth, achieving high denoising accuracy. However, these methods may ignore the underlying distribution of clean images, inducing distortions or artifacts in denoising results. This paper proposes a new perspective to treat image denoising as a distribution learning and disentangling task. Since the noisy image distribution can be viewed as a joint distribution of clean images and noise, the denoised images can be obtained via manipulating the latent representations to the clean counterpart. This paper also provides a distribution learning based denoising framework. Following this framework, we present an invertible denoising network, FDN, without any assumptions on either clean or noise distributions, as well as a distribution disentanglement method. FDN learns the distribution of noisy images, which is different from the previous CNN based discriminative mapping. Experimental results demonstrate FDN's capacity to remove synthetic additive white Gaussian noise (AWGN) on both category-specific and remote sensing images. Furthermore, the performance of FDN surpasses that of previously published methods in real image denoising with fewer parameters and faster speed. Our code is available at: https://github.com/Yang-Liu1082/FDN.git.