Chris Harrison and Scott E. Hudson of CMU’s Human Computer Interaction Institute developed a technique for simulating a 3D video conferencing experience using standard webcam equipment.
Previous research efforts have explored the use of 3D in telecommunication, and show that the additional realism can enrich the video conference experience. However, existing systems require complex sensor and cameras setups that make them infeasible for widespread adoption. We present a method for producing a pseudo-3D experience using only a single generic webcam at each end. This means nearly any computer currently able to video conference can use our technique, making it readily adoptable. Although using comparatively simple techniques, the 3D result is convincing.
The hack works by using two techniques. First, the user’s image is subtracted from the background. This allows the person’s image to be rendered in a separate plane, and perspective transformations can be applied to simulate a different camera perspective, even though the webcam is stationary. Second, the head position of the viewer is constantly tracked so that the system can adjust the received image based on where the viewer moves her head. The result is that you can move your head to the left or right and it appears as if you can see around the person you are speaking to.