The Kinect allows tracking of users without additional markers. We develop a magic mirror that generated an overlay of a video image with volume visualization from a CT volume. Such a system could be used for education of anatomy.
Detailed Project Description
The Microsoft Kinect provides a color and a depth image. Using OpenNI and PrimeSense NITE we can get the skeleton of a person standing in front of the Kinect in real-time.
We register a CT dataset to this skeleton and do an Augmented Reality overlay of the CT and the person. Using context and focus visualization we only show the CT through a virtual window, while still showing the person.
The Kinect is positioned next to a big screen. When standing in front of the system, the system acts like a magic mirror that allows seeing inside yourself. Such a system can be used for educational purposes to teach anatomy.
The system is currently a prototype. It does not show the CT of the person in front of the monitor, but a CT of another person. We use the Visible Korean Human dataset that has been provided to us by Prof. Min Suk Chung.
If you are interested in our research here you can find more information about our research on Medical Augmented Reality