Head Movement Recognition Based on Lucas-Kanade Algorithm

This paper researched some methods about head movements recognition, in allusion to some defects in these methods, we raised a new method based on the Lucas-Kande algorithm. In order to identify head movements accurately, in this paper, we detected face first, then located the nostrils, finally tracked the nostrils using Lucas-Kanade optical flow algorithm. The experimental results show that this method is robust and real-time. At the same time, it can recognize the bow and turn face aside well.