[i5heu]

Very good work!

Sadly this model really does not like nosy images that have codec compression artifacts, at least with my few test images.

[LoganDark]

I wonder if there is a de-artifacting model out there.

[earslap]

I think the company named Topaz had a photoshop plugin to remove "jpeg artifacts" - I don't know if they are using a neural model for it though.

[dimatura]

Yes, there are plenty of them. Not sure what the SOTA is, though. Similar to super-resolution, it is relatively simple to create a nearly-infinite dataset for these; pick a clean image, then introduce JPG artifacts. Then train a model to invert the process.

[LoganDark]

> Similar to super-resolution, it is relatively simple to create a nearly-infinite dataset for these; pick a clean image, then introduce JPG artifacts. Then train a model to invert the process.

Yep, exactly what I was thinking. The thing is it's hard to find enough clean images!

[numba888]

I did it some time ago. It works at low levels. Doing it in hard cases is non-trivial but possible. It's difficult to monetize that's why nobody is doing it seriously.