Miniature dioramas wouldn't be size appropriate. Apple could detect faces/cars/other common objects of ~known size and verify -- or even just dump depth map for anyone to check.

At that point you've earned the fruits of your deception, just like the tricksters who spent time doing physical photo editing.