Computer Engineering and Applications ›› 2013, Vol. 49 ›› Issue (23): 122-126.

Previous Articles     Next Articles

Mixture facial expression analysis using sparse representation

SONG Caifeng, LIU Weifeng, WANG Yanjiang   

  1. College of Information and Control Engineering, China University of Petroleum(East China), Qingdao, Shandong 266555, China
  • Online:2013-12-01 Published:2016-06-12

混合表情的稀疏表达分析

宋彩风,刘伟锋,王延江   

  1. 中国石油大学(华东) 信息与控制工程学院,山东 青岛 266555

Abstract: A quantitative description method of mixed facial expression is proposed. Based on compressive sensing, mixed facial expression is analyzed using Gabor features. Different components of mixed expression are quantitatively represented by employing membership function. Experimental results show that this method can express the composition of each basic expression in mixed expression simply and effectively.

Key words: expression analysis, compressive sensing, mixed expression, Gabor, membership function, cut set

摘要: 提出了一种混合表情的定量描述方法。基于压缩感知的理论框架,以面部特征点的Gabor小波系数为表情特征对混合表情进行了分析;利用隶属度函数定量表示混合表情中的不同组成。实验结果表明,该方法可以简单有效地表示混合表情中各基本表情的组成。

关键词: 表情分析, 压缩感知, 混合表情, Gabor, 隶属度, 截集