KinectFusion
Latest
Kinect for Windows SDK update lets developers add green screen effects
Kinect for Windows developers can now get a little more creative: Microsoft has released version 1.8 of the camera's SDK, which lets app creators produce a green screen effect by removing the background. The update also brings a new Kinect Fusion API that scans the color of an object in addition to its shape, saving some 3D modelers the trouble of creating a separate texture map. There's better scene tracking and more code samples, too. Programmers who crave the new software tricks can grab the refreshed SDK and its companion tools at the source links.
Kinect for Windows SDK gets significant update on March 18, includes Kinect Fusion and Interactions
Kinect for Windows is getting a big SDK update on March 18th to version 1.7 -- Redmond's calling it "our most significant update to the SDK since we released the first version" -- which includes the long awaited 3D object scanning application Kinect Fusion. Microsoft took to Engadget's Expand stage today to unveil the features of the SDK update, which included live demos of both Kinect Fusion and Interactions; Fusion creates live 3D models of both people and objects, while Interactions adds a whole variety of recognizable gestures to the Kinect for Windows SDK ("push-to-press buttons, grip-to-pan capabilities, and support for smart ways to accommodate multiple users and two-person interactions," says Microsoft). Microsoft' also adding code samples to its Kinect for Windows development site (CodePlex), making this the first such code from Microsoft available in an open-source channel. We'll have demo videos of the new Kinect for Windows SDK features for you as soon as we can. Follow all of Engadget's Expand coverage live from San Francisco right here!
Kinect for Windows SDK version 1.7 includes 'Kinect Fusion' 3D modeling, available March 18
An updated version of Kinect for Window's software development kit will be available from the device's official site on Monday, March 18, Microsoft's Bob Heddle announced today during Engadget's ongoing Expand event in San Francisco.Perhaps most importantly, SDK version 1.7 will include "Kinect Fusion," Microsoft's at-home 3D modeling solution that allows the Kinect to capture and fabricate real-time 3D models of people and objects. We first saw this technology in action way back in August of 2011, when the concept surfaced as a Microsoft Research project.Additionally, the new SDK also includes "Kinect Interactions," which adds support for new gestures such as "push-to-press" and "grip-to-pan," as well as "support for smart ways to accommodate multiple users and two-person interactions," according to Engadget.
Kinect Fusion-powered concept demos AR brain models for neurosurgeons (video)
Microsoft has talked up its Kinect Fusion tool since 2011, but it took some time at TechFest this week to show off how the software could be useful in operating rooms. For those who need a refresher, Redmond's solution can create 3D models of whatever an attached Kinect sensor lays eyes on, but in this instance it was leveraged to create an augmented reality experience. Using an off-the-shelf Kinect camera duct-taped to a tablet, Microsoft researchers layered a model of a brain onto a mannequin's head, making its would-be mind viewable on the slate from different angles. Ballmer and Co. reckon that neurosurgeons could use the technique to visualize what's in a patient's noggin and plan how they'll guide their scalpel. Word that Fusion would come to the Kinect for Windows SDK first surfaced last year, but Microsoft now says it'll hit the dev kit's next release, which should arrive shortly. Head past the break to catch a video of the medical concept app in action.
KinectFusion HQ creates 3D models from your surroundings
We've seen people do some pretty awesome stuff with the Kinect that expands the peripheral's applications beyond the realm of controlling games, but we've yet to see anything that matches the wide-reaching implications of Microsoft Research's KinectFusion HQ concept. Using the Kinect's dual cameras, KinectFusion HQ can construct an accurate, high-quality 3D representation of an entire room in a matter of seconds. The longer the Kinect has to look at a specific object, the more detailed that object then becomes; able to accurately scan and render millimeter differences in texture. Once the room is rendered, the model can then be interacted with in real time, which the above (oddly silent) tech demo demonstrates in increasingly wild and mind-blowing ways. Since the Kinect is typically thought of as a stationary object, it will be interesting to see how this technology may eventually be implemented into gaming, since its primary functionality relies on moving the unit around the space you're scanning.
