Unsupervised Denoising for Signal-Dependent and Row-Correlated Imaging Noise
Not on the map. Why?
Abstract (arXiv)
Accurate analysis of microscopy images is hindered by the presence of noise. This noise is usually signal-dependent and often additionally correlated along rows or columns of pixels. Current self- and unsupervised denoisers can address signal-dependent noise, but none can reliably remove noise that is also row- or column-correlated. Here, we present the first fully unsupervised deep learning-based denoiser capable of handling imaging noise that is row-correlated as well as signal-dependent. Our approach uses a Variational Autoencoder (VAE) with a specially designed autoregressive decoder. This decoder is capable of modeling row-correlated and signal-dependent noise but is incapable of independently modeling underlying clean signal. The VAE therefore produces latent variables containing only clean signal information, and these are mapped back into image space using a proposed second decoder network. Our method does not require a pre-trained noise model and can be trained from scratch using unpaired noisy data. We benchmark our approach on microscopy datatsets from a range of imaging modalities and sensor types, each with row- or column-correlated, signal-dependent noise, and show that it outperforms existing self- and unsupervised denoisers.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 33 of its 70 references that are in the map. First posted on arXiv 11 Oct 2023.
Builds on
33 of 70 references.
References · 33
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
- 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
- 2014Conditional Generative Adversarial Netscited by 2,162
- 2022Diffusion Posterior Sampling for General Noisy Inverse Problemscited by 1,416
- 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,133
- 2016Conditional Image Generation with PixelCNN Decoderscited by 862
- 2020NVAE: A Deep Hierarchical Variational Autoencodercited by 637
- 2018Noise2Noise: Learning Image Restoration without Clean Datacited by 533
- 2021Solving Inverse Problems in Medical Imaging with Score-Based Generative Modelscited by 521
- 2018Noise2Void - Learning Denoising from Single Noisy Imagescited by 364
- 2016Ladder Variational Autoencoderscited by 318
21 more
- 2019Noise2Self: Blind Denoising by Self-Supervisioncited by 276
- 2016Variational Lossy Autoencodercited by 176
- 2021Neighbor2Neighbor: Self-Supervised Denoising from Single Noisy Imagescited by 173
- 2019High-Quality Self-Supervised Deep Image Denoisingcited by 161
- 2019Noisier2Noise: Learning to Denoise from Unpaired Noisy Datacited by 126
- 2019Lagging Inference Networks and Posterior Collapse in Variational Autoencoderscited by 124
- 2023Ambient Diffusion: Learning Clean Distributions from Corrupted Datacited by 111
- 2022AP-BSN: Self-Supervised Denoising for Real-World Images via Asymmetric PD and Blind-Spot Networkcited by 93
- 2022Blind2Unblind: Self-Supervised Image Denoising with Visible Blind Spotscited by 84
- 2018A Poisson-Gaussian Denoising Dataset with Real Fluorescence Microscopy Imagescited by 71
- 2023Image Denoising: The Deep Learning Revolution and Beyond -- A Survey Paper --cited by 71
- 2016PixelVAE: A Latent Variable Model for Natural Imagescited by 70
- 2019Probabilistic Noise2Void: Unsupervised Content-Aware Denoisingcited by 51
- 2018Unsupervised Learning with Stein's Unbiased Risk Estimatorcited by 47
- 2023Solving Inverse Problems with Score-Based Generative Priors learned from Noisy Datacited by 40
- 2021Interpretable Unsupervised Diversity Denoising and Artefact Removalcited by 26
- 2021Optimal Transport for Unsupervised Denoising Learningcited by 24
- 2021Deformed2Self: Self-Supervised Denoising for Dynamic Medical Imagingcited by 15
- 2019Fully Unsupervised Probabilistic Noise2Voidcited by 14
- 2022N2V2 -- Fixing Noise2Void Checkerboard Artifacts with Modified Sampling Strategies and a Tweaked Network Architecturecited by 7
- 2023Image Denoising and the Generative Accumulation of Photonscited by 3
Built on by
Citing papers per year: 2026: 1.
1 paper.