基于Kinect的实时面部情感识别(英文)

被引:4
作者
Qi-rong MAO
Xin-yu PAN
Yong-zhao ZHAN
Xiang-jun SHEN
机构
[1] DepartmentofComputerScienceandCommunicationEngineering,JiangsuUniversity
关键词
Kinect; 情感识别; 面部表情; 实时分类; 融合算法; 支持向量机;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
目的:基于Kinect提取面部表情特征,实现对视频中面部情感实时识别。创新点:基于Kinect提出一种新颖的面部情感识别方法,充分发挥出Kinect高速易用的优势,可实时对视频序列最新连续帧中表现出的情感进行综合判定。识别方法涉及两种不同形式的表情特征,对此我们也针对性地提出基于最大置信的融合算法。方法:首先,运用Kinect中Face Tracking SDK从实时视频数据中追踪人脸、提取面部运动单元信息和特征点坐标(图3、4)。然后,将这两类特征信息并行处理,在它们各自特征通道中,特征数据经7元1-vs-1分类器组进行预识别,将得到的预识别结果存入缓存用于情感置信统计,置信度最高的即为此通道中的情感识别结果(图2)。最后,融合这两个特征通道的结果即可得到最终情感识别结果(图1)。结论:基于Kinect提取到的两种面部表情特征(面部运动单元信息和特征点坐标),提出一种新颖高效的面部情感识别方法,实现对视频中面部情感实时识别。
引用
收藏
页码:272 / 283
页数:12
相关论文
共 21 条
  • [11] Gender Classification with Visualand Depth Images. van den Hurk,Y. Journal of Women s Health . 2012
  • [12] Effectof illumination on automatic expression recognition:anovel 3D relightable facial database. Stratou,G,Ghosh,A,Debevec,P.et al. Proc.IEEE Int.Conf.on Automatic Face & Gesture Recognition andWorkshops . 2011
  • [13] Bosphorus database for3D face analysis. Arman S,Nese A,Hamdi D,et al. Biometrics and Identity Management . 2008
  • [14] Facial Action Coding System: A Technique for the Measurement of Facial Movement. Ekman P,Friesen W V. Journal of Women s Health . 1978
  • [15] Face detection,pose estimation,andlandmark localization in the wild. Zhu X X,Ramanan D. Proceedings of 2012IEEE Conference on Computer Vision and Pattern Recognition . 2012
  • [16] Bridging the Gap between Social Animal and Unsocial Machine: A Survey of Social Signal Processing. Alessandro Vinciarelli,Maja Pantic,Dirk Heylen,Catherine Pelachaud,Isabella Poggi,Francesca D’Errico,Marc Schroeder. IEEE Transactions on Affective Computing . 2012
  • [17] Facial expression and emotion. Ekman P. American Psychologist . 1993
  • [18] A survey of affect recognition methods: audio, visual, and spontaneous expressions. Z Zeng,M Pantic,GI Roisman,TS Huang. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2009
  • [19] Unsupervised facial expressions recognition and avatar reconstruction from Kinect. Seddik,B,Maamatou,H,Gazzah,S.et al. Proc.10th Int.Multi-conf.on Systems,Signals&Devices . 2013
  • [20] Graph[P]. HALEY MICHAEL B;ADISESHAN TARA A;KOGA YOSHIHITO YOTTO;GANDHI SENTHIL.中国专利:US10296626B2,2019-05-21