So for spot the difference images, I cross my eyes. The differences jump out.
HOWEVER
In this one, and I know it’s two different images, but when I tried the same trick, the image became accurately 3D. Some fuzziness, of course, but there were proper layers to the resulting depth.
I’m very curious as to the implications of how this A.I. may have been trained…
I came here to say the same thing as I too made the same discovery when I read the OP saying one side has more details on left side. I can only imagine that the prompt said side by side or something along those lines and that took it down that rat hole. This may be the worlds first spot the difference with a 3D twist.
You just gave me some ideas to try out, thanks! I’d love to see if you can ‘see’ my latest posts 3d effect.
Was curious to see how Dall-e 3 could handle making a spot the difference type image, and I’m pretty impressed if you ignore the faces it’s almost playable.
Also very interesting that the left side shows more of the right side, and the right side shows more of the left side.
Up from the bottom right corner the veg is pink in the right image.
Also most things in the top right different colours.
Sounds like entire spot-the-difference games can be generated by AI, including arts and coding.