Real-time, Continuous Hand Gestures on a Tabletop

Project Lead: Ying Yin

In pursuit of more natural modes of human-computer interaction, we developed a new interface that enables real-time natural gestural interaction with a tabletop display, in a relatively uncontrolled environment. By natural gestures, we mean continuous 3D hand gestures encountered in spontaneous interaction, rather than a set of artificial gestures chosen to simplify recognition. We also connected our gesture recognition system to Google Earth, enabling real time gestural control of a 3D map. We used the hand tracking technique developed by Robert Wang which uses a commercial webcam and requires the user to wear a colored glove. We used Hidden Markov Models for continuous gesture recognition.

Code

Contact