NAFNet

NAFNet copied to clipboard

PSNR loss and L1 Loss

1

I want to know the quantitative comparisons between PSNR loss and L1 loss. Since PSNR loss is adopted in NAFNet and HINet, I guess the PSNR loss achieves the better...

dasonglicu

模型转为onnx

6

转为onnx问题已经解决，代码已经更新。可以供给大家参考，如有侵权，可联系本人删除 https://blog.csdn.net/TF666666/article/details/125678629?spm=1001.2014.3001.5502

yanyide123

pretrained plainnet and baselinenet

2

Hello and congratulations on your ECCV22 acceptance! I wonder if it's convenient for you to share the pretrained checkpoints of PlainNet and BaselineNet in your paper, many thanks :)

YouCaiJun98

Img size

2

非常感谢你们的工作，训练时的图像尺寸必须是512吗，不同尺寸的输入会影响训练效果吗。测试时大图切片后拼接会出现块状效应，请问有什么好的消除块状的思路可以分享一下吗，感谢

juechenzp

DND数据集的结果

1

请问作者是否有(计划)提交和公开noise.visinf.tu-darmstadt.de 的结果？

madfff

CUDA out of memory

10

got the following error when running the Replicate demo (Debluring): > CUDA out of memory. Tried to allocate 5.41 GiB (GPU 0; 14.76 GiB total capacity; 9.14 GiB already allocated;...

honeytidy

Error occured during inference using HuggingFace Demo

2

The testcase used for inference: 

songkq

Performance on other datasets

1

Hi, Thanks a lot for your awesome work! And it does great help in the research work! May I ask about the performance on other low level tasks, like Deraining,...

TiankaiHang

enc_blk_nums middle_blk_num dec_blk_nums

4

Different tasks need required different settings?

Ken1256

About the folder ./datasets/GoPro/train/blur_crops

2

This is a problem when I try to run gopro.py. The program tells me that this folder already exists, but in fact it doesn't. I don't know if this is...

Tarnishedmo