基于模板的人脸点云补洞方法
|
摘要
为得到完整的三维人脸点云数据,首先提出一种在点云生成过程中自动提取三维人脸特征点的方法,然后根据三维特征点对人脸点云模型进行剖分变形。还对非刚体最近点迭代算法进行改进,加入以对应点拓扑结构为依据的剔除错误点对应点机制和优化的迭代策略,实现更快更好的配准,继而完成点云的补洞。试验结果表明,该方法对人脸点云孔洞填补效果较好,特别是在人眼、鼻子等复杂部位,补洞真实,优势明显。
To get complete point clouds, proposes a new algorithm of automatically obtain landmarks of 3D face in the generation of 3D face. With these 3D landmarks, 3D face model is deformed. Additionally, improves Nonrigid ICP algorithm including employs a new mechanism of excluding error corresponds and optimizes the strategy of iteration. Experiment results show that this method can fill holes more effective and more realistic, especially in the complicated parts such as eyes.
To get complete point clouds, proposes a new algorithm of automatically obtain landmarks of 3D face in the generation of 3D face. With these 3D landmarks, 3D face model is deformed. Additionally, improves Nonrigid ICP algorithm including employs a new mechanism of excluding error corresponds and optimizes the strategy of iteration. Experiment results show that this method can fill holes more effective and more realistic, especially in the complicated parts such as eyes.
引文
[1]Attene M,Campen M,Kobbelt L.Polygon Mesh Repairing:An Application Perspective[J].ACM Computing Surveys(CSUR),2013,45(2):3472-3476.
[2]岳杰,陆声链,孙智慧等.三维点云孔洞修补算法及在植物形态重建中的应用[J].农机化研究,2013,35(5):190-195.
[3]周亮,孙志海,吴以凡,张烨.面向人体尺寸测量系统的点云孔洞修补方法[J].杭州电子科技大学学报,2014,34(5):73-77.
[4]孙晓东.人体点云数据处理中若干问题的研究[D].北京:北京工业大学,2012:44-46.
[5]Michael Kazhdan and Hugues Hoppe,Poisson Surface Reconstruction[J].Symposium on Geometry Processing,2006,32(3):61-70.
[6]谢倩茹,耿国华.三维模型孔洞修补方法的研究[J].计算机应用研究,2013,30(10):3175-3177.
[7]王运钢,徐岩军,林海荣.基于双向切片的点云孔洞修补方法研究[J].测绘与空间地理信息.2015,38(10):218-220.
[8]晏海平,吴禄慎,陈华伟.基于径向基函数的散乱点云孔洞修复算法[J].计算机工程与设计.2014,35(4):1253-1257.
[9]罗伟正.RBF函数的孔洞修复技术研究[J].科学技术与工程,2012,12(2):404-406.
[10]张立国,王静,金梅,康乐.基于径向基函数的点云岛屿孔洞自动修补[J].高技术通讯,2016,26(2):166-172.
[11]符艳青.基于改进的径向基函数网络的3D隐式曲面重构算法研究[D].杭州:中国计量学院,2014
[12]B.Amberg,S.Romdhani,T.Vetter.Optimal Step Nonrigid ICP Algorithms for Surface Registration[J].CVPR,2007,101(1):1-8.
[2]岳杰,陆声链,孙智慧等.三维点云孔洞修补算法及在植物形态重建中的应用[J].农机化研究,2013,35(5):190-195.
[3]周亮,孙志海,吴以凡,张烨.面向人体尺寸测量系统的点云孔洞修补方法[J].杭州电子科技大学学报,2014,34(5):73-77.
[4]孙晓东.人体点云数据处理中若干问题的研究[D].北京:北京工业大学,2012:44-46.
[5]Michael Kazhdan and Hugues Hoppe,Poisson Surface Reconstruction[J].Symposium on Geometry Processing,2006,32(3):61-70.
[6]谢倩茹,耿国华.三维模型孔洞修补方法的研究[J].计算机应用研究,2013,30(10):3175-3177.
[7]王运钢,徐岩军,林海荣.基于双向切片的点云孔洞修补方法研究[J].测绘与空间地理信息.2015,38(10):218-220.
[8]晏海平,吴禄慎,陈华伟.基于径向基函数的散乱点云孔洞修复算法[J].计算机工程与设计.2014,35(4):1253-1257.
[9]罗伟正.RBF函数的孔洞修复技术研究[J].科学技术与工程,2012,12(2):404-406.
[10]张立国,王静,金梅,康乐.基于径向基函数的点云岛屿孔洞自动修补[J].高技术通讯,2016,26(2):166-172.
[11]符艳青.基于改进的径向基函数网络的3D隐式曲面重构算法研究[D].杭州:中国计量学院,2014
[12]B.Amberg,S.Romdhani,T.Vetter.Optimal Step Nonrigid ICP Algorithms for Surface Registration[J].CVPR,2007,101(1):1-8.