| 基于模型的无标识人体运动捕捉研究 |
| |
| 引用本文： | 张殿勇,苗振江.基于模型的无标识人体运动捕捉研究[J].北方交通大学学报,2014(2):42-49. |
| |
| 作者姓名： | 张殿勇 苗振江 |
| |
| 作者单位： | 北京交通大学计算机与信息技术学院,北京100044 |
| |
| 基金项目： | 国家科技支撑计划项目资助（2012BAH01F03）;国家自然科学基金资助项目（60973061）;国家“973”计划项目资助（2011CB302203）;教育部博士点基金资助（20100009110004）;北京市自然科学基金资助项目（4123104） |
| |
| 摘 要： | 无标识人体运动捕捉因其广泛的应用价值和学术价值一直是计算机视觉领域的研究热点.基于模型的无标识人体运动捕捉技术的研究目的,就是从单目或者多目摄像机图像中获得人体的运动姿态和人体模型的三维信息描述.本文从模型的构建、人体姿态估计和跟踪方法等几个方面分析了基于模型的运动捕捉的研究现状,最后阐述了该领域面临的难点问题及发展趋势.
|
| 关 键 词： | 运动捕捉 人体模型 计算机视觉 |
Research on model-based markerless human motion capture |
| |
| ZHANG Dianyong,MIAO Zhenjiang.Research on model-based markerless human motion capture[J].Journal of Northern Jiaotong University,2014(2):42-49. |
| |
| Authors: | ZHANG Dianyong MIAO Zhenjiang |
| |
| Institution: | (School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China) |
| |
| Abstract: | Makerless human motion capture （MHMC） is an active research area in the vision community for its numerous applications and academic values. The aim of modelbased MHMC is to recover human pose from one or multi camera imagery and get a 3D human model representation. This paper analyzes the stateoftheart modelbased MHMC in the aspects of human modeling, pose estimation andtracking. Finally, we conclude the paper with a discussion of this research area including the existingdifficult problems, current limitations and future directions. |
| |
| Keywords: | motion capture human model computer vision |
| 本文献已被 维普 等数据库收录！ |
