Today at the Channel 9 Live event, Microsoft Research announced the Kinect for Windows SDK Beta that enables the company’s Kinect sensor to work with Windows 7 PCs now available to download.
“Kinect SDK beta isn’t open-sourced (nor are any of its component parts). It runs on Windows 7 PCs only and is available under a custom Microsoft academic license. The SDK supports C++, C# or Visual Basic development via Visual Studio 2010, and is under 100 MB in size. The SDK was developed jointly by Microsoft Research and the company’s Interactive Entertainment Business unit.”
“Kinect SDK includes raw sensor streams from the Kinect cameras and microphone arrays for those who want to access the low-level programming interfaces. It also provides skeletal tracking capabilities for up to two people for those who want to create gesture-centric applications, as well as audio elements that’re integrated directly with Microsoft’s Speech API (SAPI).”
The SDK also includes 100 pages of documentation, reusable code samples and code walk-throughs. Microsoft is seeking feedback from the community on the SDK.
There’s no word on when Microsoft plans to take this SDK out of beta.
To mark the launch Microsoft hosted a 24 hour CodeCamp on the Microsoft campus in Redmond, Washington. The Redmond company invited 35 academics, students and other developers to campus for a Kinect SDK Code Camp. After presenting the attendees with three hours of information about the SDK, Microsoft officials gave them 24 hours to build something using the SDK.
“Kinect is just the beginning of our vision for NUI (natural user interface),” said Microsoft Research Distinguished Scientist Anoop Gupta.
Gupta said Microsoft wants to get developers to think beyond entertainment and games as to where Kinect-based applications could make sense. Gupta said he expects to see telepresence/teleconferencing, manufacturing, retail billboard, automotive and lots of other categories to be big among Kinect developers as the SDK moves from non-commercial to commercial.
Gupta noted that by using multiple Kinects, developers can build whole-body representations which, when coupled with Skype’s conferencing capabilities, potentially could make for some interesting telepresence/teleconferecing scenarios.
Features of the SDK include the following:
- Raw Sensor Streams. Developers have access to raw data streams from depth sensor, color camera sensor and the four-element microphone array. These’ll allow them to build upon the low-level streams generated by the Kinect sensor.
- Skeletal Tracking. SDK has the capability to track the skeleton image of one or two people moving within the Kinect field of view, making it possible to create gesture-driven applications.
- Advanced Audio Capabilities. Audio processing capabilities include sophisticated noise suppression and echo cancellation, beam formation to identify the current sound source, and integration with the Windows speech recognition API.
- Ease of installation. SDK quickly installs in a standard way for Windows 7 with no complex configuration required and a complete installer size of less than 100 MB. Developers can get up and running in just a few minutes with a standard standalone Kinect sensor unit widely available at retail.
- Extensive documentation. SDK includes more than 100 pages of high-quality technical documentation. In addition to built-in help files, the documentation includes detailed walkthroughs for most samples provided with the SDK.
Microsoft intends to release a commercial version of the SDK at a later date; details will be released when they’re available. The conversation is on Twitter under the hashtag #Kinect_SDK.
We’ve embedded all the Kinect for Windows SDK Quickstarts videos below:
Installing and Using the Kinect Sensor:
This video covers the basics of installing and using the Kinect sensor.
Setting Up Your Development Environment:
This video covers how to set up your development environment.
Skeletal Tracking Fundamentals:
This video covers the basics of skeletal tracking using the Kinect sensor.
This video covers the basics of reading camera data from the Kinect sensor.
Working with Depth Data:
This video covers the basics of using depth data from Kinect.
This video covers the basics of reading audio data from the Kinect microphone array, a demo adapted from the built in audio recorder. The video also covers speech recognition using Kinect. For the built in example this was based on and the speech demo in C#, check out your “My Documents\Microsoft Research KinectSDK Samples\Audio” directory.