Microsoft this morning released a new version of its Kinect for Windows software development kit, adding features including 3D modeling and new forms of user interaction. These are new features that software developers can use when creating Windows apps that make use of the Kinect sensor for controls and input.
The new features were developed by Microsoft Research. “Kinect Fusion” offers the ability to quickly make 3D models of objects and people in a room, in real time. Microsoft says in a blog post that the feature will make it possible to “create highly accurate 3-D renderings at a fraction of the price it would cost with traditional high-end 3-D scanners.”
The long-term potential for this type of technology goes much further. For example, at the newly remodeled Microsoft Envisioning Center prototyping facility, one of the company’s demonstrations uses Kinect Fusion to create a 3D model of an object and then print a replica of the object using a 3D printer.
The company is also releasing Kinect Interactions as part of the Kinect for Windows SDK update. This feature uses gives the sensor the ability to recognize whether a person’s hand is open or closed — enabling functions such as a mid-air “mouse click” for selecting content on screen. It’s an example of Microsoft using machine learning to improve the Kinect beyond its native capabilities.
Here’s a video from the Engadget Expand event with more details.