What if we could capture live video of objects or scenes from several cameras, and process it to generate 3D models and images that could then be displayed manipulated using 3D technology? For example, we could display the scene over the web on a 3D stereo screen, insert the object into a virtual world, or have a gaze-corrected video conference.

In 2008-09, we set up a new minilab to explore this area. A video featuring Vinay Vaishamayan, Shankar Krishnan and Amy Reibman, explains some of this work (primarily for a general audience).