Author Login Editor-in-Chief Peer Review Editor Work Office Work

Computer Engineering ›› 2012, Vol. 38 ›› Issue (2): 17-20. doi: 10.3969/j.issn.1000-3428.2012.02.006

• Networks and Communications • Previous Articles     Next Articles

Novel Data Conversion Method for Motion Capture

PAN Hua-wei a, MENG Qi a, GAO Chun-ming a, LEI Yuan b   

  1. (a. College of Information Science and Engineering; b. College of Mathematics and Econometrics, Hunan University, Changsha 410082, China)
  • Received:2011-07-20 Online:2012-01-20 Published:2012-01-20

一种新的运动捕获数据转换方法

潘华伟 a,孟 奇 a,高春鸣 a,雷 渊 b   

  1. (湖南大学 a. 信息科学与工程学院;b. 数学与计量经济学院,长沙 410082)
  • 作者简介:潘华伟(1972-),男,讲师、博士研究生,主研方向:虚拟现实,数字媒体;孟 奇,硕士研究生;高春鸣,教授、博士;雷 渊,副教授、博士
  • 基金资助:
    国家“863”计划基金资助项目(2007AA010404);国家科技支撑计划基金资助项目(2007BAH14B01, 2007BAH14B02, 2007 BAH14B03);湖南省自然科学基金资助项目(08JJ6041);长沙市科技计划基金资助重大项目(K1003006-61);中央高校基本科研业务费科研专项基金资助项目(531107040014)

Abstract: When calculating the rotational information of the skeleton joints, the traditional data conversion method can reduce the accuracy of the rotation data. Aiming at the problem, a data conversion method for motion capture is proposed. It establishes the human body tree structure skeleton model by using the structural relationship between their own frame human skeleton model, and designs the method of construction and decomposition to solve the three freedom rotation information of the key points, and uses this information to the drive human skeleton model. Experimental results demonstrate the effectiveness of this method.

Key words: method of construction and decomposition, freedom, skeleton model, rotational information

摘要: 传统数据转换方法在计算骨骼关节点旋转信息时,会降低旋转信息的精确度。为此,提出一种新的运动捕获数据转换方法。以树型结构建立人体骨骼模型,根据关节点自身在骨架中的结构关系,通过构造分解法求解关节点的三自由度旋转信息,利用该信息驱动人体骨骼模型。实验结果验证该方法的有效性。

关键词: 构造分解法, 自由度, 骨骼模型, 旋转信息

CLC Number: