基于视频的人体运动分析的分析-通信与信息系统专业论文.docxVIP

华中科技大学硕士学位论文 华 中 科 技 大 学 硕 士 学 位 论 文 II II Abstract 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. The thesis focuses mainly on studying the important problems of human motion analysis. How to extract the human motion region is the first step of motion analysis. In this paper, the methods of detecting human motion are discussed.Then we analyse the merits and shortcomings of these methods and verify the conclusions by the experiments. Then the paper obtains the human motion region image which can be processed in the next step. Then this paper proposes a method of detecting the lower limb joints in human walking to obtain gait parameters. Most of the existing model-based methods of human gait tracking perform are not so good in some situation as they need mannual intervention. The paper analyses the features of walking people,extracts the human body skeleton and detects the five joints of lower limbs automatically. This paper gives the experiment results of lower limb joints detection. Finally this work focuses on tracking the articulated motion and recovering the relative three dimensional motion of the figure in the form of skeleton models in monocular videos. So we can study the technology of human body action and characteristics of human motion. The following are the main tasks of this thesis: （1）Summarizing one kind of human model which suits monocular vision algorithm（2） recovering 3D skeleton from 2D joints position, and realizing it with a real skeleton model. We have completed a system called‘