An Efficient Sequence Retrieval Method Based on Motion Capture Data

摘要：

In this paper, a novel approach is presented for the motion retrieval based on the similarity of the motion data. For each sequence in the database, the Independent Component Analysis (ICA) algorithm is used to extract features, the reference vectors distribution is estimated instead of the original data.The reference vector is clustered using Iterative Self-organizing Data Analysis Techniques Algorithm (ISODATA). Finally the similarity between the query example and motion sequence in the database is measured using the Mahalanobis distance. Experimental results on the CMU database demonstrate that the proposed approach achieves good performance.

关键词： Human motion retrieval ICA ISODATA

作者: Xiaopeng Wei Shulei Yao Qiang Zhang Runian Ma

作者单位: Key Laboratory of Advanced Design and Intelligent Computing, (Dalian university) Ministry of Educati Telecommunication Engineering Institute, Air Force Engineering University, Xian, 710077, China

会议类型: 国际会议

会议名称: 2011 3rd International Conference on Computer and Automation Engineering(ICCAE 2011)(2011年第三届IEEE计算机与自动化工程国际会议)

会议地点: 重庆

会议语种:英文

页码: 384-388

在线出版日期: 2011-01-21（万方平台首次上网日期，不代表论文的发表时间）