• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2015, Vol. 37 ›› Issue (04): 802-805.

• 论文 • 上一篇    下一篇

基于改进Kinect的点分布模型面部表情特征提取

曹小峰,程实,陈翔,陈晓红   

  1. (南通大学计算机科学与技术学院,江苏 南通 226019)
  • 收稿日期:2013-05-07 修回日期:2013-09-21 出版日期:2015-04-25 发布日期:2015-04-25
  • 基金资助:

    国家自然科学基金资助项目(61202006);江苏省自然科学基金资助项目(BK2010280);南通大学自然类基金资助项目(12Z036,11Z071)

Feature extraction algorithm for facial expression based on
improved Kinect point distribution model   

CAO Xiaofeng,CHENG Shi,CHEN Xiang,CHEN Xiaohong   

  1. (School of Computer Science and Technology,Nantong University,Nantong 226019,China)
  • Received:2013-05-07 Revised:2013-09-21 Online:2015-04-25 Published:2015-04-25

摘要:

在面部表情识别中,差异性会在人体面部充分体现,各个器官的整体分布的复杂性和外界影响等因素,使得人体面部表情特征的定位非常困难。利用Kinect点分布特性,给出了一个面部表情特征提取算法,并进行了相关实验,结果表明,本方法在提取对不相关变形保持不变、噪声不敏感、类别区分度大的特征具有明显优势。

关键词: Kinect, 特征提取, 面部表情, 情感计算, 表情识别

Abstract:

In facial expression recognition system, due to the differences of faces,the complexity of organ distribution and the influence of outside factors,facial features location becomes a challenging task.Using Kinect point distribution,we present a feature extraction algorithm for facial expressions,and carry out related experiments.Experimental results show that the proposed method has significant advantages in extracting features that are with unrelated deformation,nonsensitive to noise,or have a high degree of distinction.

Key words: Kinect;feature extraction;facial expression;emotional computing