基于张量分析的表情特征提取  

Facial expression feature extraction based on tensor analysis

作　　者：孙波[1] 刘永娜[1] 罗继鸿 张迪[1] 张树玲[1] 陈玖冰[1] SUN Bo;LIU Yongna;LUO Jihong;ZHANG Di;ZHANG Shuling;CHEN Jiubing(College of Information Science and Technology, Beijing Normal University, Beijing 100875, China)

机构地区：[1]北京师范大学信息科学与技术学院,北京100875

出　　处：《计算机工程与应用》2016年第20期145-148,226,共5页Computer Engineering and Applications

基　　金：北京自然科学基金(No.4102030);中央高校基本科研业务费专项资金资助项目(No.2014KJJCA15);教育科学十二五规划课题(No.DCA140229)

摘　　要：表情识别的性能依赖于所提取表情特征的有效性,现有方法提取的表情基本上是人脸与表情的融合体,然而不同个体的人脸差异是表情识别的主要干扰因素。在表情识别时,理想情况是将个体相关的人脸特征和与个体无关的表情特征相分离。针对此问题,在三维空间建立人脸张量;然后用张量分析的方法将人脸特征与表情特征进行分离,使获取的表情参数与人脸无关。从而排除不同个体的人脸差异对表情识别的干扰。最后,在JAFFE表情数据库上验证了该方法的有效性。Facial expression feature extraction plays an important role in facial expression recognition. The expressionfeature extracted by existing methods is the combination of individual facial feature and expression feature. Facial recognitionis based on different individual facial feature, but facial expression recognition needs to find out the differences ofdifferent expressions. What is more important is individual difference will influence the facial expression recognition, andobstruct the expression reorganization rate. In an optimal situation, the related individual facial feature can be separatedduring the process of facial expression recognition. This paper presents a method that can eliminate interference of facialfeatures when recognizing the facial expression. Firstly, a three order tensor will be built. Secondly, it uses the tensor analysismethod to decompose the face feature and the expression feature into the person subspace and the expression subspacerespectively. This method can ensure that parameters of expression and face are not related. The evaluation experiment onJAFFE proves the validity of the method.

关 键 词：表情特征提取 表情识别 情感识别 张量分析 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]