基于活动外观模型的人脸表情分析与识别

被引:20
作者
左坤隆
刘文耀
机构
[1] 天津大学精密仪器与光电子工程学院光电信息技术科学教育部重点实验室
关键词
人脸表情识别(FER); 活动外观模板(AAM); 多变量统计; 人工神经网络;
D O I
10.16136/j.joel.2004.07.025
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
分析了用活动外观模板(AAM)提取的人脸表情特征来进行人脸表情识别(FER)的可行性,尝试了以此特征向量为基础的FER。根据人脸图像的特点,先用特征眼的方法定位眼睛区域,再采用AAM的优化算法获取新对象的特征,缩短了AAM方法定位新对象的优化时间,提高了定位的准确度。采用秩相关分析和非度量多维标度(nMDS)等多变量统计学方法分析得出AAM方法提取的表情特征能够很好地表达表情的变化,并构造了神经网络分类器对人脸表情图像进行识别实验,得到93.5%的识别率。
引用
收藏
页码:853 / 857
页数:5
相关论文
共 10 条
[1]  
View basedandmodulareigenspacesforfacerecognition. PentlandA,MoghaddamB,StarnerT. ProceedingsoftheIEEEConferenceonCVPR . 1994
[2]  
Statistical modeling for facial expression analysis and synthesis. Abboud B,Davoine F,Dang M. In: IEEE International Conference on Image Processing[C] . 2003
[3]  
Neural networks for pattern recognition. Bishop C. . 1995
[4]  
Face recognition using active appearance models. Edwards G J,Cootes T F,Taylor C J. In: Proc . European Conf . Computer Vision . 1998
[5]  
Comparison between geometry-based and Gabor-wavelets-based facial expression recognition using multi-layer perceptron. Zhang Z,Lyons M,Schuster M,et al. In: Proc . Third IEEE Conf . Face and Gesture Recognition . 1998
[6]  
Advances in active appearance models. Edwards G J,Cootes T F,Taylor C J. In: Proc International Conference on Computer Vision . 1999
[7]  
Active appearance models. Cootes T F,Edwards G J,Taylor C J. In: 5 th European Conference on Computer Vision . 1998
[8]  
Coding, analysis interpretation, recognition of facial expressions. Essa I,Pentland A. IEEE Transactions on Pattern Analysis and Machine Intelligence . 1997
[9]  
Facial expression recognition using model-based feature extraction and action parameters classification. Huang C L,Huang Y M. J Visual Comm and Image Representation . 1997
[10]  
Coding facial expressions with Gabor wavelets. Lyons M,Akamatsu S,Kamachi M,et al. In: Proc . Third IEEE Conf . Face and Gesture Recognition . 1998