Smart TV interaction system using face and hand gesture recognition

Abstract

In this paper, a face and hand gesture recognition system which can be applied to a smart TV interaction system is proposed. Human face and natural hand gesture are the key component to interact with smart TV system. The face recognition system is used in viewer authentication and the hand gesture recognition in control of smart TV, for example, volume up/down, channel changing. Personalized service such as favorite channels recommendation or parental guidance can be provided using face recognition. We show that the face recognition detection rate is about 99% and the face recognition rate is about 97% by using DGIST database. Also, hand detection rate is about 98% at distance of 1 meter, 1.5 meter, and 2 meter, respectively. Overall 5 type hand gesture recognition rate is about 80% using support vector machine (SVM). © 2013 IEEE.