AAM-Candide结合的快速表情动画合成模型  

AAM-Candide Combining Quick Expression Animation Model

作　　者：蒋巍[1] 卢娜 Jiang Wei;Lu Na(Art College,Xi’an University of Science and Technology,Xi’an 710054,China;College of Economics and Management,Xi’an Aeronautical University,Xi’an 710077,China)

机构地区：[1]西安科技大学艺术学院,西安710054 [2]西安航空学院经济管理学院,西安710077

出　　处：《科技通报》2017年第9期191-194,共4页Bulletin of Science and Technology

摘　　要：针对现有的算法在表情动画合成中自动化程度不高、生成效果不理想的问题,本文提出了一种基于AAM-Candide结合的快速表情动画合成模型。首先利用Candide-3算法获取面部表情参数-AU,并通过确定人脸的Candide-3模型来计算AU,将参数提取问题转化为一个非刚体的人脸跟踪问题,然后利用彩色图对人脸关键点进行2D定位,通过上下限规范实现Candide-3和三维数据点云之间的对应点集。实例仿真结果表明,本文提出的改进算法,对惊讶、悲伤、嘟嘴、开心等4种表情得到了较好的动画合成效果。Aiming at the problem that the existing algorithm is not highly automated in the expressionanimation synthesis,this paper presents a fast expression animation synthesis model based on AAMCandidecombination.First,the Candide-3algorithm is used to obtain the facial expression parameter-AU,and the AU is calculated by determining the Candide-3model of the face.The problem of parameterextraction is transformed into a non-rigid face tracking problem,and then the color Point for2Dpositioning,through the upper and lower limits to achieve Candide-3and three-dimensional data pointcloud between the corresponding point set.The simulation results show that the improved algorithmproposed in this paper has a good effect on animation,such as surprise,sadness,beep,happy and so on.

关 键 词：表情动画 AAM模型 Candide-3模型 人脸跟踪 关键点定位 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]