You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thanks for you great work. I have a problem of the SRCNN model results. It seems that SRCNN performed poorly than bicubic method in all datasets. It may be somewhat strange that deep-learning-based method performed worse than interpolation. Could you provide any insight for this phenomenon?
I guess that SRCNN utilized the stride LR SM for input while bicubic do not?
The text was updated successfully, but these errors were encountered:
Thanks for you great work. I have a problem of the SRCNN model results. It seems that SRCNN performed poorly than bicubic method in all datasets. It may be somewhat strange that deep-learning-based method performed worse than interpolation. Could you provide any insight for this phenomenon?
I guess that SRCNN utilized the stride LR SM for input while bicubic do not?
The text was updated successfully, but these errors were encountered: