cszn/DnCNN
A deep CNN model achieving state-of-the-art image denoising through residual learning, applicable to Gaussian denoising, JPEG deblocking, and super-resolution.
Not currently ranked — collecting fresh signals.
star history
DnCNN implements a residual learning approach for image denoising using deep convolutional neural networks. The model demonstrates state-of-the-art performance on various image restoration tasks including Gaussian denoising, JPEG artifact removal, and super-resolution. It has been ported to multiple frameworks (PyTorch, Keras/TensorFlow, MatConvNet) with compatible model parameters across implementations. The approach uses batch normalization merged into convolutional layers for efficient inference.
Frequently asked
- What is cszn/DnCNN?
- A deep CNN model achieving state-of-the-art image denoising through residual learning, applicable to Gaussian denoising, JPEG deblocking, and super-resolution.
- Is DnCNN open source?
- Yes — cszn/DnCNN is an open-source project tracked on heatdrop.
- What language is DnCNN written in?
- cszn/DnCNN is primarily written in MATLAB.
- How popular is DnCNN?
- cszn/DnCNN has 1.7k stars on GitHub.
- Where can I find DnCNN?
- cszn/DnCNN is on GitHub at https://github.com/cszn/DnCNN.