Thank you for your contribution to the joint super-resolution and motion deblurring task. I have some questions regarding your reported comparison results.
How many images did you use for testing on the GoPro dataset? As far as I know, the GoPro test set contains 1,111 images. When I tested using the provided pretrained code, I obtained results that were significantly better than those reported in your paper. What could be the reason for this discrepancy?
Thank you for your contribution to the joint super-resolution and motion deblurring task. I have some questions regarding your reported comparison results.
How many images did you use for testing on the GoPro dataset? As far as I know, the GoPro test set contains 1,111 images. When I tested using the provided pretrained code, I obtained results that were significantly better than those reported in your paper. What could be the reason for this discrepancy?