计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (18): 148-153.

• 图形、图像、模式识别 • 上一篇    下一篇

面向内容的三维人体运动检索技术研究综述

刘贤梅,赵  丹   

  1. 东北石油大学 计算机与信息技术学院,黑龙江 大庆 163318
  • 出版日期:2012-06-21 发布日期:2012-06-20

Survey on content-based retrieval of 3D human motion data

LIU Xianmei, ZHAO Dan   

  1. School of Computer & Information Technology, Northeast Petroleum University, Daqing, Heilongjiang 163318, China
  • Online:2012-06-21 Published:2012-06-20

摘要: 三维人体运动数据检索技术是计算机动画领域的研究热点,系统分析了当前基于内容的三维人体运动数据检索技术的研究现状。分析比较了基于运动数值特征、基于运动能量特征和基于语义信息特征三种运动数据的特征表示方法;从局部相似性度量和全局相似性判断两个角度阐述了运动相似性匹配算法;总结目前基于内容的人体运动数据检索技术中存在的问题,探讨了未来开展基于语义的检索技术所需的工作。

关键词: 人体运动数据, 运动检索, 特征表示, 相似性匹配

Abstract: 3D human motion data retrieval is an important and interesting problem in computer animation. This paper presents a comprehensive survey on state of the art of content-based motion data retrieval. Three representations of motion features including numerical feature, energy feature and semantic feature are compared; local similarity measure and global similarity matching as two parts of similarity judging algorithm are analyzed; some problems of content-based retrieval and future work of semantic-based retrieval are discussed.

Key words: human motion data, motion retrieval, feature representation, similarity matching