A novel approach to hand gesture contour detection based on GVF Snake model and skin color elliptical model

摘要

Hand gesture contour detection is the key technology of gesture recognition based on computer vision. In this paper, the GVF Snake model and the skin color elliptical model is proposed to extract the contour of the static hand gesture. The skin color elliptical model is used to calculate the field of the hand and the initial contour. Hand gesture contour is detected by the GVF Snake model based on the initial contour. Experiment results shows that the approach is obviously superior to the GVF Snake model or the skin color elliptical model in aspects of accuracy and real-time.