表情识别研究 Sources of facial expressions Facial Expression Recognition 表情识别是基于视觉信息对脸部的运动以及脸部特征的形变进行分类。表情识别主要包括脸部的获取、脸部特征提取和特征的分类。主要的工作集中于脸部特征的提取和特征的分类。其中更侧重于分类问题的研究。考虑表情在表情空间中不是简单的线性可分的。所以考虑使用非线性降维和统计模型相结合的方法进行表情的分类。 Facial expression recognition deals with the classification of facial motion and facial feature deformation into abstract classes that are purely based on visual information. The facial recognition consists of face acquisition, facial feature extraction and facial expression classification. Our works focus on the facial feature extraction and facial expression classification and emphasize particularly on the research of classification. The facial expression is not classified by using linear classifier in the facial expression space. So we use methods of the nonlinear and the statistic to recognize facial expression. Sources of facial expressions Generic facial expression recognition framework