Neural Affine Grayscale Image Denoising

TL;DR

Neural AIDE introduces a neural network-based grayscale image denoising method that learns an affine mapping per pixel, enabling both supervised training and adaptive fine-tuning directly on noisy images, outperforming many existing methods.

Contribution

The paper presents a novel neural network framework that learns pixel-wise affine mappings for denoising, with an estimated loss function based solely on noisy data, allowing effective fine-tuning.

Findings

Outperforms recent state-of-the-art denoising methods on benchmark datasets.

Enables adaptive fine-tuning to match noise variance during testing.

Addresses limitations of patch-level supervised learning in denoising.

Abstract

We propose a new grayscale image denoiser, dubbed as Neural Affine Image Denoiser (Neural AIDE), which utilizes neural network in a novel way. Unlike other neural network based image denoising methods, which typically apply simple supervised learning to learn a mapping from a noisy patch to a clean patch, we formulate to train a neural network to learn an \emph{affine} mapping that gets applied to a noisy pixel, based on its context. Our formulation enables both supervised training of the network from the labeled training dataset and adaptive fine-tuning of the network parameters using the given noisy image subject to denoising. The key tool for devising Neural AIDE is to devise an estimated loss function of the MSE of the affine mapping, solely based on the noisy data. As a result, our algorithm can outperform most of the recent state-of-the-art methods in the standard benchmark datasets. Moreover, our fine-tuning method can nicely overcome one of the drawbacks of the patch-level supervised learning methods in image denoising; namely, a supervised trained model with a mismatched noise variance can be mostly corrected as long as we have the matched noise variance during the fine-tuning step.