As a static background. As I expected, it made the skull into a semi-3D object.

iOS has been able to do this for a few years now. Take a picture of someone, press and hold on the subject, and you can make it a sticker. Now it automatically picks the subject (in this case, the skull) and tries to render it 3D. It has enough data to extrapolate what is behind the skull since it is a repeating pattern. I would imagine a more capable AI would look around the skull and realise it’s a repeating pattern, and then render the pattern entirely. Then it could find a 3D image of a skull that looks exactly like that, delete the 2D one, and place a 3D one in its place. Then it could maybe even animate it.