search

KupynOrest / DeblurGAN

Image Deblurring using Generative Adversarial Networks
Other
2.48k stars 513 forks source link

How to generate the results showed in the paper? #102

Open KWang1998 opened 5 years ago

KWang1998 commented 5 years ago

I just used the default parameters to train the model with the given dataset and test it with my own images and images from GoPro dataset, but it couldn't work well as what the paper shows. After deblurring, the images still have some blur and have some grids at the background. I just used the default parameters. So, would you please tell me how to generate the results as good as which showed in paper?

chenyuyi94 commented 5 years ago

I just used the default parameters to train the model with the given dataset and test it with my own images and images from GoPro dataset, but it couldn't work well as what the paper shows. After deblurring, the images still have some blur and have some grids at the background. I just used the default parameters. So, would you please tell me how to generate the results as good as which showed in paper? hi,i didnt train the dataset,i only using the code of [python test.py --dataroot /.path_to_your_data --model test --dataset_mode single --learn_residual] then ,didt show the result .could you please tell me how to generate the result image,thank you