you must be thinking of something else. kinect has two cameras, one for laser IR particle-filter type stuff, and then an rgb camera. it isn't stereo vision with two regular 2d cameras, however, doing displacement math. it is looking at the grid of dots projected and seeing where the dots moved in order to create a depth map.