Research On Face Tracking Technologies Based On Movement Prediction

With the development of computer science and technology, the computer applications have entered into every aspect of human life, and play a very important role. In which, artificial intelligence is very hot research field, and also the foundation of the artificial computer. It is always a very important subject in the research of artificial intelligence that the computer can recognize the object as a man, and it is also the first step the computer understanding the world.This paper is focusing on the trend of face tracking technology based on motion prediction. Face tracking includes not only the detection and location of faces but also tracking it by using motion prediction algorithm in real-time environment. After analyzing the existing detection methods, we introduce skin-color detection method to detect face position roughly. Then template matching and border detection method is used to detect accurately. Combine the two results and get the precise location. Experiment showed that the above algorithm can get more precise position of faces than other algorithms. But, the complexity of template matching algorithm needs further research. In the tracking process motion prediction method is used to estimate the face position in future frames to reduce to using of detection algorithm and to improve the real-time adaptation of face tracking system. In the tracking strategy algorithm we create a status conversion of tracked faces so as to improve algorithm robustness when the faces are in sideways position or obscured in short time. We proposed a prediction method based on linear. It is fast and precise. Specific face tracking is realized by using the combination of specific face template updating and specific face skin-color.Face tracking algorithm based on motion prediction can specific characters in image accurately. It has wide range of application scenarios. Experiments show that the face tracking algorithm can finish tracking human face fast and accurately.