yu4u/noise2noise
An unofficial and partial Keras implementation of "Noise2Noise: Learning Image Restoration without Clean Data"
Implements flexible noise models (Gaussian, text insertion, random-valued impulse) and supports both SRResNet and UNet architectures with pluggable loss functions (MAE, L0), enabling comparative training between noise-to-noise and noise-to-clean paradigms. Built on Keras/TensorFlow with configurable training pipelines that decouple source and target noise distributions, allowing investigation of how different noise types affect restoration performance across image restoration tasks.
1,092 stars. No commits in the last 6 months.
Stars
1,092
Forks
235
Language
Python
License
MIT
Category
Last pushed
Aug 12, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/yu4u/noise2noise"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
CAREamics/careamics
A deep-learning library for denoising images using Noise2Void and friends (CARE, PN2V, HDN...
rgeirhos/texture-vs-shape
Pre-trained models, data, code & materials from the paper "ImageNet-trained CNNs are biased...
NICALab/SUPPORT
Accurate denoising of voltage imaging data through statistically unbiased prediction, Nature Methods.
cabooster/DeepCAD-RT
DeepCAD-RT: Real-time denoising of fluorescence time-lapse imaging using deep self-supervised learning
jaewon-lee-b/lte
Local Texture Estimator for Implicit Representation Function, in CVPR 2022