微表情自动识别综述  

Survey on Automatic Micro Expression Recognition Methods

作　　者：贲晛烨[1,2] 杨明强[1] 张鹏[1] 李娟[3] 

机构地区：[1]山东大学信息科学与工程学院,济南250100 [2]南京理工大学高维信息智能感知与系统教育部重点实验室,南京210094 [3]山东大学心理健康教育与咨询中心,济南250100

出　　处：《计算机辅助设计与图形学学报》2014年第9期1385-1395,共11页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61201370);教育部博士点基金(20120131120030);中国博士后科学基金(2013M530321);中国博士后科学基金第7批特别资助(2014T70636);山东省博士后创新项目专项资金(201303100);高维信息智能感知与系统教育部重点实验室(南京理工大学)基金(30920140122006);山东大学自主创新基金(2012GN043;2012DX007;2012ZD039;IFW09108)

摘　　要：微表情是一种持续时间仅为1/25～1/5 s非常快速的表情,它表达了人试图压抑与隐藏的真正情感.文中简要分析了面部行为代码系统,探讨了人工微表情识别方法的实验研究;对现有的微表情自动识别的数据库从采集情况、微表情的产生情况以及用途进行了对比总结,重点综述了现有的微表情自动识别研究进展,从识别结果、所采用的技术、优缺点等方面进行了对比总结,给出了评估方法,论述了微表情线索测谎的典型应用;最后探讨了该领域值得进一步研究的问题和可能的发展方向.Micro expression is a fast facial movement, which usually lasts for 1/25 to 1/5 second. It reveals a real emotion that people try to suppress and conceal. In this paper, facial action coding system （FACS） is firstly analyzed briefly, and then the experimental study of artificial micro expression recognition is explored. Secondly, the acquisition, generation and usage of micro expressions are comparatively summarized based on the existing automatic micro expression recognition database. Thirdly, the comparative studies of recognition results, technologies, advantages and disadvantages with an emphasis on automatic micro expression recognition methods are also reviewed. Fourthly, the evaluation measures of micro expression are discussed. Fifthly, a typical application of using micro expression clues to deception detection is addressed. Finally, several issues worthy of further study and possible directions of development are concluded.

关 键 词：微表情 微表情自动识别 测谎 面部行为代码系统 微表情识别评估方法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]