Joint self-supervised blind denoising and noise estimation

TL;DR

This paper introduces a self-supervised method for blind image denoising that jointly predicts clean images and noise distribution without needing clean training data, especially useful for biomedical imaging.

Contribution

It presents a novel joint neural network approach for blind denoising and noise estimation that outperforms existing methods and is computationally efficient.

Findings

Outperforms state-of-the-art self-supervised denoising algorithms on biomedical datasets.

Effectively captures noise distribution with synthetic data.

Framework is simple, lightweight, and practical.

Abstract

We propose a novel self-supervised image blind denoising approach in which two neural networks jointly predict the clean signal and infer the noise distribution. Assuming that the noisy observations are independent conditionally to the signal, the networks can be jointly trained without clean training data. Therefore, our approach is particularly relevant for biomedical image denoising where the noise is difficult to model precisely and clean training data are usually unavailable. Our method significantly outperforms current state-of-the-art self-supervised blind denoising algorithms, on six publicly available biomedical image datasets. We also show empirically with synthetic noisy data that our model captures the noise distribution efficiently. Finally, the described framework is simple, lightweight and computationally efficient, making it useful in practical cases.