HAT icon indicating copy to clipboard operation
HAT copied to clipboard

Published 20 hours ago • verified publisher icon XPixelGroup

Results are larger but not improved?

Open AIisCool opened this issue 2 years ago • 1 comments

Input: image

Output: image

AIisCool avatar May 21 '22 07:05 AIisCool

@AIisCool The provided pretrained models aim to perform High-Quality Image Super-Resolution, for which can not deal with other degradations like compression or blurring. We would consider training a HAT model that can handle multiple degradations, but no model is available yet currently. This project might help Real-ESRGAN.

chxy95 avatar May 22 '22 10:05 chxy95