“With a single image and a small amount of annotation,” writes researcher Kevin Karsch, “our method creates a physical model of the scene that is suitable for realistically rendering synthetic objects.” Fascinating:
Check out the project site for much more detailed info. [Via Zorana Gee]
Amazing indeed. Can’t wait for Adobe to buy them, and put this in Photoshop/After Effect!
Hopefully planning for this eventuality is the reason why we now have 3D tools in Ps.
This, along with the motion de-blur technology, is going to make Photoshop CS6 great! 😉