Head Pose Estimation for Video Image and Applied Research
Chen Duan-sheng · Mind and Computation · 2011
In video applications,Accurate estimation of head pose face not only conducive to capture facial close-up,but also contribute to the further attention of researchers.A relatively practical head pose estimation in video is presented Based on the video head tracking technology.Mean Shift and Particle Filter algorithm are used for real-time head tracking;The Locality preserving projection method is secondly used to reduce the dimension of the image;SVM classifier is finally used to estimate pose.The experiments show that this method can effectively track the human head image and better estimate the head pose.