首页|基于多角度LBP特征的三维人脸性别分类

基于多角度LBP特征的三维人脸性别分类

扫码查看
原文链接
  • 国家科技期刊平台
  • NETL
  • NSTL
  • 万方数据
  • 维普
人脸性别分类是一个富有挑战的研究方向,目前的研究尚不完善.本文提出一种三维人脸的性别分类方法,首先对数据集进行局部区域最近邻点迭代算法(Iterative closest point,ICP)匹配,自动实现人脸正向姿态校正；对数据集人脸统一做俯仰角度的旋转,从不同视角上提取基于深度缩略图的多角度LBP (Local binary patterns)特征；再由支持向量机(Support vector machine,SVM)分类器完成训练分类.该方法在CASIA数据库上实验,对全库中性表情人脸进行性别分类,可以得到最高98.374％的正确率.
3D Facial Gender Classification Based on Multi-angle LBP Feature
Facial gender classification is a challenging topic, and it's still not perfect until now. In this paper, we propose a series of methods of gender classification based on three-dimension faces. Automatic front-pose adjustment is needed through local region iterative closest point (ICP) registration firstly; then we do pitching rotating and extract muni-angle LBP features from depth thumbnail map in different viewing angles; at last, we use support vector machine (SVM) classifier to do training and prediction. This algorithm has been experimented on CASIA database, and for the neutral faces in this database, we can get a highest correct classification rate of 98.374%.

3D face, gender classification, local region iterative closest point (ICP), depth thumbnail map, multi-angle LBP

赵海英、杨一帆、徐正光

展开 >

新疆师范大学计算机科学技术学院 乌鲁木齐830054

北京科技大学自动化学院 北京100083

三维人脸 性别分类 局部区域最近邻点迭代算法(Iterative closest point,ICP) 深度缩略图 多角度LBP

国家自然科学基金国家自然科学基金973前期计划专项课题

60973064611630442010CB334709

2012

10.3724/SP.J.1004.2012.01544

自动化学报
中国自动化学会 中国科学院自动化研究所

自动化学报

CSTPCDCSCD北大核心EI
影响因子：1.762
ISSN：0254-4156
年,卷(期)：2012.38(9)
  • 5
  • 1