| 面向视频序列表情分类的LSVM算法 |
| |
| 引用本文： | 徐文晖,孙正兴. 面向视频序列表情分类的LSVM算法[J]. 计算机辅助设计与图形学学报, 2009, 21(4) |
| |
| 作者姓名： | 徐文晖 孙正兴 |
| |
| 作者单位： | 南京大学计算机软件新技术国家重点实验室,南京,210093;南京大学计算机软件新技术国家重点实验室,南京,210093 |
| |
| 基金项目： | 国家自然科学基金，国家高技术研究发展计划(863计划)，教育部新世纪优秀人才支持计划 |
| |
| 摘 要： | 为了提高基于视频序列的表情识别精度,在KNN-SVM算法的基础上提出局部SVM分类机制,并将其用于视频序列中的表情分类.对于一个待分类的几何特征样本,首先在训练集中寻找该样本的k个近邻样本,然后根据这k个近邻样本和待分类样本的相似度信息,重新构建局部最优的SVM分类决策超平面,用来对该几何特征样本进行分类.在Cohn-Kanade数据库中的对比实验表明,该分类器有效地提高了表情分类的精度.
|
| 关 键 词： | 表情识别 局部SVM KNN-SVM 几何特征 |
Facial Expression Recognition from Image Sequences with LSVM |
| |
| Xu Wenhui,Sun Zhengxing. Facial Expression Recognition from Image Sequences with LSVM[J]. Journal of Computer-Aided Design & Computer Graphics, 2009, 21(4) |
| |
| Authors: | Xu Wenhui Sun Zhengxing |
| |
| Affiliation: | State Key Laboratory for Novel Sof tware Technology;Nanjing University;Nanjing 210093 |
| |
| Abstract: | In order to improve the accuracy of video based facial expression recognition,we propose a local SVM based on KNN-SVM algorithm,applied in facial expression recognition. For a geometric feature test sample,we first select k nearest neighboring training samples. A local optimal SVM decision hyper-plane is rebuilt based on the similarity between the test sample and the k neighboring samples for classifying the geometric feature test sample. Four different classifiers,KNN,SVM,KNN-SVM and LSVM,were compared,and... |
| |
| Keywords: | KNN-SVM |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
