REAL TIME HAND TRACKING AND GESTURE RECOGNITION FOR INTERACTIVE VIRTUAL HAND CONTROL

Abstract

Objects tracking, captured from the real word, is an area of research that has attracted a lot of attention lately, for its potential regarding the interaction between man and machines. The ability to track hand movement provides new opportunities for human computer interaction and also hand gesture recognition, in real time, and the video stream plays a significant role in controlling a virtual hand in computer generated reality applications. On the current, digital image processing applications this represent a difficult task. The present paper aims to present a new method for human hand control in virtual environments, by eliminating the need of an external device currently used for hand motion capture and digitization. A first step in optimizing gesture recognition will be to detect and track in an efficiently manner the human hand from a video stream, then detect the gesture only from a region of a picture and use the result to control a virtual hand into a virtual environment.