If I have an color deviation image that doesn't know the ground truth, can the network correct it?
ToChangeBetter opened this issue · 1 comments
ToChangeBetter commented
If I have an color deviation image that doesn't know the ground truth, can the network correct it?
mahmoudnafifi commented
You can pre-trained models without training. Just keep in mind that you need to provide the model with multiple images as described in the paper/readme.