DmitryUlyanov / deep-image-prior

Image restoration with neural networks but without learning.

Home Page:https://dmitryulyanov.github.io/deep_image_prior

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

"super-resolution.ipynb" , can I use this code to upscale image without quality reduction?

mohammad-saber opened this issue · comments

Hi, Thanks for sharing your great code.

Assume that I have a "png" image file with size of 128 by 128. I want to increase its size into "384 by 384" without reduction in the image quality. How can I use "super-resolution.ipynb" for this purpose?

I found that "super-resolution.ipynb" loads an image of "zebra" and creates image "HR_out_np" in the code, having almost the same size of the original image file. Also, it creates variable "result deep prior" which has the same size of original image file.

What if I want to increase image size?

In a case that I don't have ground truth for high resolution image, can I still use your code to create high resolution image from low resolution image?

Thanks!