基于改进肌肉模型的三维人脸自适应参数配置  

ADAPTIVE PARAMETER CONFIGURATION OF 3D FACE BASED ON IMPROVED MUSCLE MODEL

作　　者：周悦 刘凯[1] Zhou Yue;Liu Kai(College of Electrical Engineering,Sichuan University,Chengdu 610065,Sichuan,China)

机构地区：[1]四川大学电气工程学院,四川成都610065

出　　处：《计算机应用与软件》2022年第6期127-132,140,共7页Computer Applications and Software

基　　金：国家自然科学基金面上项目(61473198);四川省科技厅重点研发项目(2018GZ0198);成都市科技局技术创新研发项目(2018-YFYF-00029-GX);四川大学自贡市校地科技合作重点研发项目(2018CDZG-12)。

摘　　要：三维人脸表情合成技术是动画、电影等诸多领域的热点,Waters肌肉模型是常见的表情合成模型。针对传统肌肉模型的不足,提出一种基于改进模型的三维人脸自适应肌肉参数配置方法。在线性肌模型中加入斥力弹簧和边约束,解决肌肉收缩力大时网格出现尖锐突出和拓扑结构被破坏的问题。根据一般人脸和特定人脸的特征映射关系,实现肌肉参数的自适应配置。通过实验证明了该方法合成的表情更具有真实感,且参数配置精度高,时间效率提升。The technology of 3D facial expression synthesis is a hot spot in animation,film and many other fields.Waters muscle model is a common expression synthesis model.In view of the shortcomings of the traditional muscle model,we propose an adaptive muscle parameter configuration of 3D face based on the improved model.In the linear muscle model,we added the repulsion spring and edge constraint to solve the problem of sharp mesh and topological structure damage caused by the large muscle contraction force.According to the feature mapping relationship between general face and specific face,the adaptive configuration of muscle parameters was realized.The experimental results show that the expression synthesized by this method is more realistic,and the parameter configuration accuracy is high,and the time efficiency is improved.

关 键 词：人脸表情 肌肉模型 弹簧模型 边约束 自适应参数 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]