Gesture Recognition using a Range Camera
Michael Boelstoft Holte, Thomas Baltzer Moeslund · 2007
Abstract. This paper introduces use of range information acquired by a CSEM SwissRanger SR-2 camera for one and two arms gesture recognition. The range data enables motion detection and 3D representation of gestures. Motion is detected by double difference range images and filtered by a hysteresis bandpass filter. Gestures are represented by shape contexts in the form of spherical histograms. This representation allows a simple matching by binary comparison of the histogram bins. Although this approach is still on a early state, results indicate successful recognition of a set of commonly used gestures. 1