Journal of Systems Engineering and Electronics ›› 2011, Vol. 33 ›› Issue (8): 1891-1895.doi: 10.3969/j.issn.1001-506X.2011.08.40

• 软件、算法与仿真 • 上一篇    下一篇

基于统计学习的逆向运动学实现方法

瞿师1, 吴玲达1,2, 魏迎梅1, 李松1, 冯晓萌1   

  1. 1. 国防科学技术大学信息系统工程重点实验室, 湖南 长沙 410073; 2. 装备指挥技术学院, 北京 101400
  • 出版日期:2011-08-15 发布日期:2010-01-03

Inverse kinematics based on statistical learning

QU Shi1, WU Ling-da, WEI Ying-mei1, LI Song, FENG Xiao-meng   

  1. 1. Information System Engineering Key Lab, National University of Defense Technology, Changsha 410073, China; 
    2. School of Equipment Command Technology, Beijing 101400, China
  • Online:2011-08-15 Published:2010-01-03

摘要:

基于统计学习的思想,提出一种逆向运动学实现方法。角色动画运动数据维数较高,各维度之间存在相关性,直接对其分析计算复杂度高。该方法基于高斯过程隐变量模型对运动数据降维,将高维运动数据映射到二维隐空间,对隐空间数据进行聚类,寻找样本运动数据的典型姿态,典型姿态张成的子空间保留了样本运动数据的主要特征和规律。结合末端约束,对典型姿态进行加权优化,得到满足末端约束的最佳姿态。实验表明,该方法取得了较好的效果。

Abstract:

An inverse kinematics solution based on statistic learning is presented. Because of the high -dimension of character animation motion data and correlation lying in various dimensions, it is a very hard work to analyze and compute directly. The motion data are mapped from high-dimensional space to two-dimensional latent space based on Gaussian process latent variable models (GPLVM). Then, the representative poses of virtual characters are found out by clustering the motion data in latent space, which can expand a subspace that contains the primary characters and disciplinarians of training data. Finally, the weight of representative poses is optimized combined with constraints on the end effectors, and the optimized pose is obtained. The experiments show that the proposed method obtains a better effect.