I like this:

The software program processes the video stream data captured by the camera and tracks the locations of the colored markers (visual tracking fiducials) at the tip of the user’s fingers using simple computer-vision techniques. The movements and arrangements of these fiducials are interpreted into gestures that act as interaction instructions for the projected application interfaces. The maximum number of tracked fingers is only constrained by the number of unique fiducials, thus SixthSense also supports multi-touch and multi-user interaction.

That means the software should, eventually, be able to provide, for example, a playable virtual piano keyboard and, in conjunction with other sofware, translate the finger and key movements into sounds. I assume that, in time, the software and hardware components will be able to track the speed and abruptness of one's finger movements, translating that into the analogus key stroke if one had a real key under one's finger.