Please use this identifier to cite or link to this item: http://hdl.handle.net/1959.14/119613
90 Visitors97 Hits0 Downloads
Hand gesture recognition based on segmented singular value decomposition
International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (14th : 2010) (8 - 10 September 2010 : Cardiff, UK)
Setchi, Rossitza; Jordanov, Ivan; Howlett, Robert J. and Jain, Lakhmi C.. Knowledge-based and intelligent information and engineering systems : 14th International Conference, KES 2010, Cardiff, UK, September 8-10, 2010, proceedings, part II, p.214-223
The increasing interest in gesture recognition is inspired largely by creating a system which can identify specific human gestures and using gestures to convey information or control devices. In this paper we present a novel approach for recognizing hand gestures. The proposed approach is based on segmented singular value decomposition(SegSVD) and considers both local and global information regarding gesture data. In this approach, first singular vectors and singular values are evaluated together to define the similarity of two gestures. Experiments with hand gesture data prove that our approach can recognize gestures with high accuracy.