We changed the URL from https://petapixel.com/2021/08/30/googles-new-ai-photo-upscal..., which points to this. Both articles point to this one, which had a recent and related thread:
High Fidelity Image Generation Using Diffusion Models - https://news.ycombinator.com/item?id=27858893 - July 2021 (19 comments)
Here's an unofficial copy of the code: https://github.com/Janspiry/Image-Super-Resolution-via-Itera...
There might be hope for Star Trek:Deep Space Nine yet!
Finally something to bring back to life those 16x16 px Windows 3.1 icons.
If we start with multiple source images that are "small" (by some definition of small) perturbations of each other and upscale them, what can be said about the results?
Some of the images in the article seemed to be high-res images that where downscaled to low-res (and it makes sense to see how the upscalling process changes the original), but wouldn't that make it easier for the ML to revert the downscaling process rather than taking an original low-res photo and upscale it?
It is pretty impressive/crazy how well CDM and SR3 work together to go from 32x32 to 256x256 e.g. the Irish Setter. How could the algos possibly know the lighter coloring (due to breed or lighting) between the dog's eyes?! It's basically inventing pixels