Kinect Fusion demonstrated at Microsoft Research TechFest, coming soon to SDK


Revealed in November as a future addition to the Kinect for Windows SDK, Kinect Fusion made a big impression at the annual TechFest event hosted by Microsoft Research this week in Redmond, Washington.

Kinect Fusion pulls depth data that is generated by the Kinect for Windows sensor and, from the sequence of frames, constructs a highly detailed 3-D map of objects or environments. The tool averages readings over hundreds or even thousands of frames to create a rich level of detail.

The Kinect Fusion project, shown during TechFest 2013, enables high-quality scanning and reconstruction of 3-D models using just a handheld Kinect for Windows sensor.
Kinect Fusion, shown during TechFest 2013, enables high-quality scanning and reconstruction of
3-D models using just a handheld Kinect for Windows sensor and the Kinect for Windows SDK.

“The amazing thing about this solution is how you can take an off-the-shelf Kinect for Windows sensor and create 3-D models rapidly,” said Shahram Izadi, senior researcher at Microsoft Research Cambridge. “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.”

When scanning smaller objects, you also have the option to simply move the object instead of the sensor.

The Cambridge researchers and Kinect for Windows team collaborated closely on Kinect Fusion to construct a tool that can enable businesses and developers to devise new types of applications.

“This has been a wonderful example of collaboration between Microsoft Research and our product group,” said Kinect for Windows Senior Program Manager Chris White. “We have worked shoulder-to-shoulder over the last year to bring this technology to our customers. The deep engagement that we have maintained with the original research team has allowed us to incorporate cutting edge research, even beyond what was shown in the original Kinect Fusion paper.”

“This kind of collaboration is one of the unique strengths of Microsoft, where we can bring together world-class researchers and world-class engineers to deliver real innovation,” White added. “Kinect Fusion opens up a wide range of development possibilities—everything from gaming and augmented reality to industrial design. We’re really excited to be able to include it in a future release of the Kinect for Windows SDK.”

Kinect for Windows team

Key Links

Comments (9)

  1. nice can't wait for next release of the SDK, just one question, when?

  2. Erik Demontigny says:

    Just like Tomasz asked… When ???   :o)

  3. Manuel says:

    Finally! I have to say i bought the Kinect for Windows just because of the promise of Kinect Fusion being released!

    When?????

  4. Bob says:

    I've been using the SDK and looking forward to new features. To add to the other comments, it would be useful to have some idea, however vague, of when we might be seeing the next upgrade and what it might contain. If its over 3 months away, we'd know where we stand and at least some uncertainty would be removed.

  5. very happy to see some word out of the Kinect folks on the sdk

    Same question as the others – I'm kind of holding back on some dev work because it looks like they've dealt with a lot of the things I would need to implement (not just the fusion stuff, but the gesture libraries as well), based on the October/Nov video.  Hope it's coming out soon!

  6. Taylor says:

    Now if only they would make Kinect for xbox work with windows…

  7. Hello all,

    Thank you for your interest in Kinect Fusion and the upcoming release of the Kinect for Windows SDK. We have no additional details at this time, but we will be sharing some news next Saturday, March 16, at Engadget Expand, so please stay tuned. We’ll post something that day to this blog, or you can see our director Bob Heddle’s talk live http://www.engadget.com/…/ticket-info

    As always, please follow us on twitter.com/KinectWindows and http://www.facebook.com/KinectForWindows

    Thank you,

    Kinect for Windows team

  8. Ari Heinonen says:

    Maybe we can 3D-scan furnitures, objects, user faces etc. and import (mesh+uv map) data to Second Life?

  9. Jesse DePinto says:

    So excited about this! Once the Kinect 2 comes out, will we be able to just plug it in with code written from the Kinect 1 and hit "run"?

Skip to main content