Gesture recognition for augmentative human computer interaction

Abstract

Augmentative and alternative communication has always suffered from difficulties of equipment set-up and interaction. A three-dimensional input device is presented which is based on a non-contact movement sensor. In this way the movement of the head or limb can be tracked and converted into the movement of a PC mouse. The advantage of the noncontact approach is further increased by the incorporation of adaptive gesture recognition thus allowing user defined movement interpretation and click strategies