Study On Human Body Motion Based On Monocular Vision

Video human motion analysis is an important research area which combines various techniques. The focus of the research is in detecting and tracking figures in monocular or multicular videos, capturing and reconstructing the human motion, and then describing and interpreting the human behavior. This research has various applications in human animation, computer game, virtual reality and augmented reality, human-computer interaction video surveillance, sports video analysis, computer-aided clinical diagnosis, and so on.This thesis focuses on the problems of capturing 2D human posture in videos. This work investigates on tracking the articulated motion and evaluating the relative three dimensional motion of the figure in the form of articulated skeleton chains in monocular videos. Combine the digital human model with the virtual environment;visual simulation and rational evaluation on ergonomics could be preformed in a Virtual-Reality circumstance. Prototype manufacturing could be eliminated;improper designs avoided;product development cycle shorted;product development cost greatly reduced. In addition, putting the 3D digital human model into virtual workplace, many ergonomics problems could be solved such as workers attainable region estimating, standard manipulation establishing and working security analyzing.After full understanding correlation research work, this article completes the research work mainly includes: (1) Summarized one kind of human model which suits monocular vision algorithm(2)Utilizing plane template to calibrate camera, avoided the non-linear superfluous iterative calculation and simplified the calculation process greatly.(3)According to the movement point of discontinuity, divided the human body movement sequence into certain subsequences, thus effectively eliminated the ambiguous disturbance.(4) Under the perspective projection camera model, the 3D human skeleton model posture can be obtained by using 2D data.(5)The accuracy of the 3D human posture depends upon human's visual judgment completely, therefore the reconstruction precision is very difficult to appraise. Having designed an experiment, we have confirmed the precision of this article algorithm.