Improvements and new features of the Kinect for Windows SDK 1.8

This March, Microsoft launched the Kinect for Windows SDK 1.7 update, including the gesture recognition Kinect Interactions and the real time 3D modeling Kinect Fusion two new technologies. Yesterday, Microsoft released the 1.8 version of the Kinect for Windows SDK, the new version of the SDK was a minor update, but a lot of updates to the examples in the Kinect Developer Toolkit, including new support for HTML5, new background removal capabilities, Kinect Fusion's color capture function, the robustness of the Kinect fusion tracker, and the multiple Kinect fixed real time three-dimensional modeling features.

Here's a look at the specifics of these new features:

New Background Removal Example:

Back to remove technology and become a "blue screen-like" technology, that is, the characters from the background to pull out, and then replace the background to other scenes of the technology, this in such as weather forecasts, as well as some films are often used in shooting. In real life can be widely used in advertising photography, virtual reality games, training simulation and other related to the virtual reality of the technology. In earlier versions, we used the Kinect sensor's deep image data to determine whether a pixel in a picture was playerindex to determine if the pixel was a character to pick a character from the background, but the earlier version had a more severe use of burrs, This can be seen from my previous introduction of depth image data written in a screenshot of the effect. The new API provides a new background removal technology, which shows that compared to the previous version, the character edge smoothing has been greatly improved.

The color capture function of Kinect Fusion

The Kinect fusion feature, introduced in version 1.7, can help us to model objects quickly and in three dimensions. In the 1.8 SDK, the Kinect Fusion application interface scans scene color and depth information, captures three-dimensional scene colors and three-dimensional models, and obtains texture mappings from scans for the grid, a feature that provides scan-complete high-fidelity three-dimensional models, Includes color, CAD and other applications for full color three-dimensional printing or game-creation of exact three-dimensional resources. This feature makes it much better and more realistic for applications such as virtual fitting to dress and body fitting.

Enhanced Kinect Fusion Tracking robustness

The algorithm simplifies the scene scanning. With this update, the Kinect Fusion is a better way to keep the scene locked while the camera position is moving, and to achieve a more reliable and consistent scan.

HTML Interaction Example

This example shows the implementation of the Kinect enabling button, simple user participation, and streaming based on the HTML5 background. Developers can implement the Kinect enabling user interface with any programming language based on HTML5 and JavaScript to integrate the Kinect for Windows into existing solutions.

See more highlights of this column: http://www.bianceng.cnhttp://www.bianceng.cn/Programming/net/