NAFNet
Visit ToolNAFNet is an open-source image restoration model that excels in deblurring, denoising, and super-resolution. It achieves state-of-the-art results without relying on nonlinear activation functions.
At a glance
Trending
NAFNet is an open-source image restoration model that excels in deblurring, denoising, and super-resolution. It achieves state-of-the-art results without relying on nonlinear activation functions.
Trending
About
NAFNet, or Nonlinear Activation Free Network, is an innovative open-source image restoration model developed by Megvii Research. It challenges conventional approaches by demonstrating that nonlinear activation functions are not necessary for achieving state-of-the-art performance in image restoration tasks. The model is highly efficient and delivers superior results in image deblurring, denoising, and stereo image super-resolution. For instance, it surpasses previous state-of-the-art methods in PSNR on datasets like GoPro and SIDD, often with significantly reduced computational costs. NAFNet provides pretrained models and detailed instructions for installation and quick start, making it accessible for researchers and developers to implement and test its capabilities.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending