Kinect for Windows (not Xbox 360) is about to get a whole lot more useful. At TechFest in Redmond, Washington, Microsoft Research showed off Kinect Fusion, a new feature that will allow the Kinect for Windows sensor to scan real objects and create 3D digital models out of them.
Microsoft's Kinect for Windows blog says that Kinect Fusion can average out "hundreds of thousands of frames" of 3D data to "construct a highly detailed 3-D map of objects or environments." Microsoft Research Cambridge's senior researcher Shahram Izadi explains:
"Normally when you think of Kinect, you think of a static sensor in a living room. But with Kinect Fusion, we allow the user to hold the camera, explore their space, and rapidly scan the world around them."
Kinect Fusion's technology was originally shown as a concept at E3 2010, but the feature never materialized when the sensor launched later that year. At E3 2011, Kudo Tsunoda, Microsoft's creative director for Kinect Games, touted object scanning as the future yet again with Kinect Fun Labs. Tsunoda said that things like plush toys and model cars could be digitized into games and used in different virtual worlds, but again, we didn't see any games take advantage of the object scanning feature.
Outside of a few user-created hacks and object-detecting in NBA Baller Beats, object scanning fell off the list of Kinect priorities. Microsoft hasn't provided any formal reason as to what happened, but It's possible that the Kinect for Xbox 360's limitations held it back.
Whatever the reason, we're excited for the possibilities that Kinect Fusion will bring. Although it's only been announced for Kinect for Windows, we imagine a version for Kinect 2 and the next Xbox isn't far behind.