Just within a couple of days after the release, Kinect is likely to gain attention for many developers and Hackers. Thanks to Hector Martin who started this feat. Recently, Oliver Kreyl from UC Davis made use of the Depth Camera to capture 3D scenes. On his website, he claimed that he didn't use any of Hector Martin's code, but the "magic incantations" that need to be sent to the Kinect to enable the cameras and start streaming. The 3D reconstruction code is entirely written from scratch in C++, using his own Vrui VR toolkit for 3D rendering management and interaction.
The captured scene is not full 3D, it only renders maximum to what the camera can see. As on the image below, when the scene is rotated, it cannot display half of the "other side" which the Kinect camera cannot see. But it does a stunning job in calculating the depth of items.
Hope there is more to come in the future. I can imagine using 2 Kinect sensors on 2 sides facing one another creating a full 3D scene. Hardware for only 300 bucks, and you even get a game.
Source : Oliver Kreyl