Distribution Conditional Denoising: A Flexible Discriminative Image Denoiser

TL;DR

This paper presents a flexible, discriminative image denoiser using multi-task learning and conditional training, achieving state-of-the-art results across various noise types and levels.

Contribution

Introduces a novel distribution conditional denoising framework that generalizes across noise types and levels using affine transforms conditioned on noise parameters.

Findings

Achieves state-of-the-art performance on Gaussian and Poisson noise.

Generalizes fixed noise level denoisers to multiple noise levels.

Uses distribution of noise parameters during training for robustness.

Abstract

A flexible discriminative image denoiser is introduced in which multi-task learning methods are applied to a densoising FCN based on U-Net. The activations of the U-Net model are modified by affine transforms that are a learned function of conditioning inputs. The learning procedure for multiple noise types and levels involves applying a distribution of noise parameters during training to the conditioning inputs, with the same noise parameters applied to a noise generating layer at the input (similar to the approach taken in a denoising autoencoder). It is shown that this flexible denoising model achieves state of the art performance on images corrupted with Gaussian and Poisson noise. It has also been shown that this conditional training method can generalise a fixed noise level U-Net denoiser to a variety of noise levels.