Stefano Carrino, Elena Mugellini, Omar Abou Khaled, and Rolf Ingold
Subtle gesture recognition, Wearable computing, Human Computer Interaction
In this paper we present a novel interaction modality based on the use of economic, subtle gestures. Gestures are performed with small effort in front of a camera worn on the user’s wrist. This sensing method, called WristEye, can recognize both deictic and iconic gestures. Frames acquired from the arm-worn camera are elaborated in order to extract two kinds of information: in which direction the arm is moving and the number of fingers in the image. Input gestures are recognized, elaborated and sent to the feedback system. Finally two applications of the WristEye system have been developed and tested. With a gesture recognition time inferior to 150 ms, real-time interaction is achieved.
Important Links:
Go Back
