There's usually nothing left of the original image. But sometimes a specific image pops up in the dataset more often and gets overtrained, which is why you can get a pretty close copy of the Starry Night from vanilla SD. But yeah, it's not tracing.
DekkerNSFW
joined 1 year ago
Sadly, AI isn't as good with sparse data like vertices and bones, so most attempts to use AI on 3D stuff is via NERFs, which is closer to a "photo" you can walk around in than to an actual 3D scene.
It can be useful, in that many neurodivergent people suffer from similar reactions from neurotypicals, and a similar lack of understanding and support. But it does nothing for the specific problems of any one type of neurodivergence, no. Consider it more of a catch-all, like how LGBT is also grouped for having similar struggles in society despite the T being quite different from the LGB part.