Axis transformation and exponential map based on human motion tracking

Abstract

Human motion tracking is a special important research domain in the field of computer vision. It is widely used in security monitoring, video compression, video editing, sports training and so on. Therefore, this paper utilizes the parameters of exponential map to express all parts of the body posture. It mainly consists of the following parts: establish the objective function based on the video image, seek to reach the minimum parameters by iteration method and utilize the minimum parameters to obtain the place of all parts of the body posture in the current frames. The experimental results indicate the presented algorithm can track human motion effectively and accurately. © 2013 Springer-Verlag.