dcscn-super-resolution icon indicating copy to clipboard operation
dcscn-super-resolution copied to clipboard

Published 20 hours ago • verified publisher icon jiny2001

Residual learning at the output

Open batraamul opened this issue 1 year ago • 3 comments

We input a size x image and get size X^2 at the output in iteration. Which image is considered as HR in calculating residual for tunning the network parameters? As currently we do not have an HR image available. Secondly, in batch processing what is the difference between true image and input image? How true images are generated from the input image patches?

batraamul avatar Feb 25 '24 14:02 batraamul