南京邮电大学学报（自然科学版）2018，Vol.38Issue(1)：60-65,6.DOI:10.14132/j.cnki.1673-5439.2018.01.007
基于人脸表情和语音的双模态情感识别
Bimodal emotion recognition based on facial expression and speech
摘要
Abstract
In the area of future artificial intelligence,the emotion recognition of the computers will play a more important role.For the bimodal emotion recognition from facial expression and speech,a feature fusion method based on sparse canonical correlation analysis is presented.Firstly,the emotion features from facial expression and speech are respectively extract.Then,the parse canonical correlation analysis is used to fuse the bimodal emotion features.Finally,the K-nearest neighbor classifier is used for emotion recognition.The experimental results show that the bimodal method based on the sparse canonical correlation analysis can obtain better recognition rate than the speech and the facial expression with single modal.关键词
人脸表情/语音/双模态情感识别/稀疏典型相关分析Key words
facial expression/speech/bimodal emotion recognition/sparse canonical correlation analysis分类
信息技术与安全科学引用本文复制引用
闫静杰,卢官明,李海波,王珊珊..基于人脸表情和语音的双模态情感识别[J].南京邮电大学学报（自然科学版）,2018,38(1):60-65,6.基金项目
国家自然科学基金(61501249)、江苏省自然科学基金(BK20150855)、江苏省重点研发计划(BE2016775)和南京邮电大学引进人才科研启动基金(NY214143)资助项目 （61501249）
