"Visual perception is the ability to interpret information and surroundings from the effects of visible light reaching the eye. The resulting perception is also known as eyesight, sight, or vision (adjectival form: visual, optical, or ocular)."
Visual perception, Wikipedia
We have had optical character recognition (OCR) for years now. Reading can be perceived as an important ability of our visual perception and we have more or less cracked it. What we don't have is the ability to acquire 3D models from a simple 2D camera, just like our eyes can. Or at least I thought so.
An article that I read recently has pushed me to google for such type of software. Interestingly enough just a few weeks ago I was thinking about such a system and its (big) importance to AI development. Google provided an article immediately, and it dated from 2009! Creating 3D models with a simple webcam (w/ Video) described a system just like the one that I envisioned and provided a nice video. Here are also some videos on YouTube: video1 (ProFORMA: Probabilistic Feature-based On-line Rapid Model Acquisition ), video2 (Interactive Model Reconstruction with User Guidance (ProFORMA)).
"GTA in your own city !!!"
Cuervaud, 1 year ago
This was one of the comments on YouTube and it just gives one idea of what this technology can do. We might be limited in our capabilities to see and remember but a computer can store and process a video of a flight over a city, thus we can easily end up with a good 3D model of an area.
There is surely a lot of material on the subject and I think I'm only scratching the surface. Only the future will tell what the best applications of such a technology will be but I know that this is one step further, and it is a big one.
Edit 01.06.2011: And the commercial applications appear... Sony Sets Its Sights on Augmented Reality :) What are the odds?