Kinect Fusion is a program that creates a full 3D model of your surroundings as you scan it with a hand-held Kinect. Now there is news that it will be included in the next version of the SDK.
Kinect Fusion is another product of Microsoft Research. It makes use of some clever 3D algorithms to piece together a full 3D model from hand-held scans using the Kinect. It does this in real time, which opens up lots of possibilities for both serious applications and games. The original demo of Kinect Fusion contains lots of examples of virtual fluids being thrown over objects and flowing correctly using the 3D model to determine how they should flow.
The really clever part of Kinect Fusion is that the more you scan a region the better the model becomes, so you can start off with a rough model and end up with something that is better than a single scan could produce.
This is all very exciting, but of little use if you can't get your hands on it. There is an open source implementation of the algorithm available, but this only works with the open source drivers for Kinect and not with the official SDK. At this year's Build, Kinect Fusion was demonstrated and made available for programmers to play with. Now the Kinect for Windows Blog has announced that Kinect Fusion will be included as a tool in the next version of the SDK.
This is the sort of cutting edge technology that will keep the Microsoft SDK ahead by allowing commercial exploitation of the technology. There are no more details at the moment, but let's hope that instant 3D is soon available for us all to use.