Just a random idea I had at some point. Don’t know enough about the technology to know how one would go about it, but …
So “Magic Eye” pictures (a.k.a. autostereograms) are those 3D images that you have to stare at a book to get. A few folks can see the images really quickly, some can only see them after staring awhile, and apparently there are quite a few who are never able to see the things.
It’s kind of an interesting effect though if you can get it to work: Hiding information in the background noise in such a way that when you are able to interpret it you can trick each eye into interpreting it differently.
I think one of the things that makes it tough to pick out the 3D image is that there are no obvious visual cues. It seems to me that in a lot of cases with normal vision the human brain is picking out the edges of things, or contrasts and gradations of solid colors that define shape. But those things are absent in autostereograms, so it’s tougher to pick them out.
It occurred to me though that what would happen if you ran together a string of similar autosterograms to produce an animation? The consistency of the 3D image against a changing background might make it easier to spot the image than with a non-animated version.
Voila! I’ve just invented 3D animation without special glasses!
Well, not quite. Apparently someone else already thought of it.
I don’t know if the image portrayed on wikipedia is exactly the best example though. ‘Twer it me, I probably would’ve made the background as a more randomized image (like basic static), rather than a sweeping colored pattern which seems to distract from the 3D image.
Aside from the wikipedia article though I’m having trouble tracking down other good examples of animated autostereograms. Seems to me there should be some small creative sector devoted to them though: Cartoons maybe. Or segments of horror film where a random background (foliage or TV static for instance), kind of becomes 3D and leaps out at the viewer. Or possibly some sort of video game.
I also wonder if there might be ways to color the 3D objects, or have the background pattern be somehow meaningful in the context of the 3D scene it’s self, rather than just the standard splatter painting effect.