文献类型：期刊文献

英文题名：Accurate 3D nose tip localization using orientation invariant surface descriptor

作者：Ju, Quan[1,2]; Wang, Yingfeng[1]

第一作者：Ju, Quan

通讯作者：Ju, Q.

机构：[1] College of Computer and Information Engineering, Henan University of Economics and Law, Zhengzhou, 450002, China; [2] Department of Computer Science, The University of York, York, YO105DD, United Kingdom

第一机构：河南财经政法大学计算机与信息工程学院

通讯机构：[1]College of Computer and Information Engineering, Henan University of Economics and Law, Zhengzhou, 450002, China|[1048412]河南财经政法大学计算机与信息工程学院;[10484]河南财经政法大学;

年份：2012

卷号：6

期号：10

起止页码：144-152

外文期刊名：International Journal of Digital Content Technology and its Applications

收录：EI(收录号：20122615185636);Scopus(收录号：2-s2.0-84862749432)

语种：英文

外文关键词：Digital communication systems - Software engineering

摘要：In this paper, we propose a 3D nose tip localization approach based on a 3D orientation-invariant surface/shape descriptor to accurately identify the nose tip. A binary neural network technique - k-Nearest Neighbour Correlation Matrix Memories algorithm (kNN CMM) is used to search and match the feature patterns of the nose tip. Our approach achieves 99.96% in nose tip identification in experiments performed on the FRGC dataset (4950 3D faces).