NAFNet
NAFNet copied to clipboard
megvii-research
The state-of-the-art image restoration model without nonlinear activation functions.
Hi, In **NAFNet**, inappropriate dependency versioning constraints can cause risks. Below are the dependencies and version constraints that the project is using ``` addict future lmdb numpy opencv-python Pillow pyyaml...
How you prepare training dataset for raw data denoising? Just like PMRID? Do I need k-sigma transformation like PMRID?
您好，我用提供的NAFNet-GoPro-width64.pth在GoPro测试集上做了测试，PSNR结果只有33.08dB.,这与论文中的结果不符，但是提供的图片结果是和论文中相符的。请问提供pre-trained model是否提供错了，可以提供PSNR=33.71dB的checkpoint吗？
Thank you for your awesome code! I am hoping you might open-source the log files you have from training. Maybe the training and validation loss as a function of epoch...
Added necessary code changes and merged with repo https://github.com/XPixelGroup/BasicSR for inference of the below paths python basicsr/demo.py -opt options/test/SIDD/NAFNet-width64.yml --input_path ./demo/noisy.png --output_path ./demo/denoise_img.png python basicsr/demo.py -opt options/test/REDS/NAFNet-width64.yml --input_path ./demo/blurry.jpg --output_path...
Thanks for your great work. I have two issues, can nafnet models(Deblurring or Denoising) be used for video enhancement or just for image!? and do you develop models that do...
does nafnet have a model for debluring and denoising simultaneously?
there is no 'create_model' method in basicr !!! ImportError: cannot import name 'create_model' from 'basicsr.models' (/usr/local/lib/python3.10/dist-packages/basicsr/models/__init__.py)
Metadata
Owner
Metadata
The state-of-the-art image restoration model without nonlinear activation functions.