Fully Articulated Hand Tracking with the Microsoft Kinect v2 Sensor
At a recent trip to the Microsoft Future Decoded conference in London, members of the Tech Office team saw a demo of some exciting research being conducted by Microsoft into hand-pose / hand-gesture recognition using the Kinect v2 sensor that we’ve previosuly been using for full-body gesture recognition and tracking.
The demo showed real-time tracking of a single hand using the depth-sensor capabilities of the Kinect which appears extremely powerful and capable. No code samples are yet available for us to use, but the research certainly looks like it is not too far away from being made available for re-use. The applications for Psychological studies could prove very interesting, particularly as the Kinect hardware is so inexpensive and Martyn Atkins from the tech team already has much experience of programming using Kinect technologies.
We present a new real-time articulated hand tracker which can enable new possibilities for human-computer interaction (HCI). Our system accurately reconstructs complex hand poses across a variety of subjects using only a single depth camera. It also allows for a high-degree of robustness, continually recovering from tracking failures. However, the most unique aspect of our tracker is its flexibility in terms of camera placement and operating range.
For more information, please see the Microsoft Research project page.