MIXing It Up: the Kinect for Windows SDK
Back in February at TechForum, Craig Mundie, Microsoft’s chief research and strategy officer, and Don Mattrick, president of Microsoft’s Interactive Entertainment Business (IEB), announced that Microsoft Research and IEB would release a non-commercial Kinect for Windows software development kit this spring. Addressing a growing body of academic researchers and enthusiasts who are anxious to build applications employing Kinect’s natural user interface, Mundie and Mattrick offered tantalizing promises of access to Kinect’s system capabilities, including audio, system APIs, and direct control of the Kinect sensor.
Today at the MIX developer conference in Las Vegas, Scott Guthrie, corporate vice president of the Microsoft .NET Developer Platform, unveiled three key features of the upcoming Kinect for Windows SDK: robust skeletal tracking, advanced audio capabilities, and XYZ depth camera. He also announced the launch of a new website for the SDK, where you can subscribe to a newsfeed and be notified as soon as the SDK is available for download.
Our hope is that this “starter kit” for application developers will make it easier for the academic research and enthusiast communities to create even richer experiences by using Kinect technology. Here are a few details on each of the SDK’s ground-breaking NUI features:
- Robust skeletal tracking will provide high-performance capabilities for tracking the skeletal image of one or two people moving within the Kinect field of view.
- Advanced audio will enable great sound capabilities by using a four-element microphone array with sophisticated acoustic noise and echo cancellation. The advanced audio will also include beam formation to identify the sound source and integration with the Windows speech recognition API.
- XYZ depth camera will provide a standard color camera stream along with depth data indicating the distance of the object from the Kinect camera. This will give developers access to the raw data and enable the creation of novel interfaces by using the unaltered data.
As is often the case, the sum of these features is greater than the parts. By combining the audio, depth, and image data, developers will have great opportunities to build deeper NUI experiences. And just to give his audience a taste of what these features will enable, Guthrie demoed a version of the WorldWide Telescope that you can interact with by using gestures—a feature built on the SDK platform.
MIX was an ideal setting for announcing the new SDK features, as this annual gathering brings together developers, designers, UX experts, and business professionals who are creating some of the most innovative consumer sites on the web and beyond. The SDK feature announcements will be highlighted to the academic research community this week at the Microsoft Research Software Summit in Paris.
So, it’s onward and upward with the Kinect for Windows SDK. We’re confident that this non-commercial SDK will fuse the work of Microsoft Research with the creativity of the academic research and enthusiast communities to deliver NUI applications that will revolutionize our relationship with computers.
- Kinect for Windows SDK beta
- Windows Phone, Kinect Exemplify New Usage Scenarios and Device Capabilities at MIX11” Press Release
—Tony Hey, Corporate Vice President, Microsoft Research Connections