摘要
从实时图像中识别面部表情和推断情感是一个极具挑战性的研究课题。文章介绍了一种根据视频图像对面部表情进行实时识别的方法。使用ASM方法和改进的L-K光流算法进行面部特征定位和特征跟踪,提取的面部特征位移作为支持向量机分类器的输入。实验证明,支持向量机和特征跟踪的方法能很好地识别面部表情。
Enabling computer systems to recognize facial expressions and infer emotions from them in real time presents a challenging research topic. In this paper, real time approach to recognize facial expression is present in live video.We employ ASM method and an improved L-K optical flow algorithm to perform face localization and feature tracker.The facial feature displacements are used as input to a Support Vector Machine classifier. Our experiments demonstrate the effectiveness of a support vector machine and feature tracking approach to facial expressions recognition.
出处
《微型电脑应用》
2010年第6期8-10,1,共3页
Microcomputer Applications
关键词
面部表情
支持向量机
特征跟踪
特征位移
Facial Expression
Support Vector Machines
Feature Tracking
Feature Displacements.
