Efficient Blind-Spot Neural Network Architecture for Image Denoising

Image denoising is an essential tool in computational photography. Standard denoising techniques, which use deep neural networks at their core, require pairs of clean and noisy images for its training. If we do not possess the clean samples, we can use blind-spot neural network architectures, which...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Honzátko, David, Bigdeli, Siavash A, Türetken, Engin, Dunbar, L Andrea
Format Paper Journal Article
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 25.08.2020
Subjects
Artificial neural networks
Computer architecture
Computer Science - Computer Vision and Pattern Recognition
Computer Science - Learning
Neural networks
Noise reduction
Performance enhancement
Pixels
Software
Training
Online AccessGet full text

Cover

More Information
Summary:Image denoising is an essential tool in computational photography. Standard denoising techniques, which use deep neural networks at their core, require pairs of clean and noisy images for its training. If we do not possess the clean samples, we can use blind-spot neural network architectures, which estimate the pixel value based on the neighbouring pixels only. These networks thus allow training on noisy images directly, as they by-design avoid trivial solutions. Nowadays, the blind-spot is mostly achieved using shifted convolutions or serialization. We propose a novel fully convolutional network architecture that uses dilations to achieve the blind-spot property. Our network improves the performance over the prior work and achieves state-of-the-art results on established datasets.
ISSN:2331-8422
DOI:10.48550/arxiv.2008.11010