三维人脸表情识别研究与进展  

The Researches and Progress of 3D Facial Expression Recognition

作　　者：严政[1] 潘志庚[1] 

机构地区：[1]杭州师范大学数字媒体与人机交互研究中心,浙江杭州311121

出　　处：《杭州师范大学学报（自然科学版）》2016年第1期88-94,共7页Journal of Hangzhou Normal University(Natural Science Edition)

基　　金：国家科技支撑计划项目(2013BAH24F00);国家科技支撑计划项目(2015BAK04B05)

摘　　要：随着人机交互与情感计算的快速发展,人脸表情识别已经成为研究热点.二维人脸表情图像对于姿态变化与光照变化不具备鲁棒性,为了解决这些问题,研究者们使用三维人脸表情数据进行表情分析研究.本文在前人工作的基础上,对三维人脸表情识别中的对齐与跟踪、表情数据库、特征提取等方面进行综述.指出人脸表情识别的热点与趋势以及存在的局限,并对未来发展进行了展望.With the fast development of human-computer interaction, the recognition of facial expression has already become the research focus. 2D facial expression images do not have the robustness of posture change and illumination variation. In order to solve these problems, researchers use 3D facial expression data to analyze expression. Basing on the prophase work, this survey summarizes the registration and tracing, expression data and feature extraction in 3D facial expression recognition. The focus, trend and limitation of facial expression recognition are put forward, and the future development is discussed.

关 键 词：情感计算 表情识别 三维表情识别 表情特征提取 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]