人脸表情自动识别技术研究进展_张家树.pdfVIP

40　 3 　　　　　　　 Vo.l 40　No. 3 2005 6 　　　　　　JOURNAL OF SOUTHWEST JIAOTONG UN IVERSITY　　　　　　 Jun. 2005 　　:0258-2724(2005)03-0285-08 1 1 2 1 1 张家树 , 　陈　辉, 　李德芳, 　罗小宾, 　夏小东 (1. , 610031;2. , 610031) 　:, , . 3、 .. 、、. , . , 、 、. :;;; :TP391. 41　　:A R esearch A dvances in A utom atic Fac ial E xpression R ecognition 1 1 2 1 1 ZHAN G J ia-shu , 　CHEN H u i , 　LI D e-f ang , 　LUO X iao-bin , 　X IA X iao-d ong (1. Signal and Inform ation ProcessingKey Lab. of Sichuan Province, SouthwestJiaotong University, Chengdu 610031, Ch ina;2. School ofHum an ities and Social Science, Southw est Jiaotong University, Chengdu 610031, China) Ab stract:A s a challenging interdiscipline of biologic feature recognition and affection calculation, the technique of automatic recognition of facial expression (facial expression automatic recognition system, FEARS)develops quickly driven by dem ands of various applications. Nevertheless, fully autom atic facial expression recognition system sw ith acceptable robustness have notyet come forth due to the great difficulties. The three key procedures for automatic recognition of facial expression are facial mi age detection, location and extraction of facial features and emotion classification, and great advances have been achieved on the three fields. The remaining problem s include mi proving the robustness of facial recognition algorithms, precise and pertinence of identification of facial features, separation of rigid facial actions and recognition of three-dmi ensional facial expressions. To build the data bases of facial expressions and application research are also mi portan