A $3 gesture recognizer for devices with 3D acceleration sensorsIn “A $3 gesture recognizer: simple gesture recognition for devices equipped with 3D acceleration sensors,” Sven Kratz and Michael Rohs of Deutsche Telekom Laboratories describe a simple but robust gesture recognition system for input devices featuring 3D acceleration sensors. It makes use of an algorithm designed to be implemented quickly and which is device-independent and does not require any special toolkits or frameworks.
Relying solely on simple trigonometric and geometric calculations, when users tested and evaluated their system a correct gesture recognition rate of 80% was achieved when using a set of 10 unique gestures for classification.
They believe their method requires significantly less training data than other gesture recognizers and is thus suited to be deployed and to deliver results rapidly.
Log-in or register for a free account to get unlimited articles and full access to EDN.