基于GPU的三维人脸数据动态线性快速修复
|
摘要
基于CUDA和Open GL,根据三维人脸的深度与灰度数据构造三维点云模型,在更短的时间内显示具有逼真效果的三维人脸模型,并根据传统三维人脸点云显示对数据采集要求较高的缺陷,提出了一种用于修补低质量三维采集数据中出现的空洞现象的动态线性修复方法并使用GPU为算法进行了加速。该方法从人脸数据集中读取点云数据,逐行扫描寻找点云数据中出现的空洞并使用符合人脸本质特征的线性插值方式修补空洞。实验证明,该方法能得到远好于传统点云显示的效果,且经过GPU加速后显示效率得到大幅提高。
In this paper,the modeling efficiency is highly improved by using GPU( CUDA),Open GL to construct the 3D pointcloud model,and the model is realistic. To resolve the problem of traditional point cloud display which requires high quality database,this paper proposes a dynamic linear recovering method to recover the holes which occur in 3D pointcloud model if it is based on low quality 3D data,and then use GPU to accelerate this process.This pointcloud data is inputed from the 3D face database and scanned progressively for holes occuring in the pointcloud,and the linear interpolating method which is consistent with human facial intrinsic feature to recover the holes is also used. The experiments show that the display effect could be far better than that the traditional point cloud display by our method,and the display efficiency can be greatly improved by using GPU acceleration.
In this paper,the modeling efficiency is highly improved by using GPU( CUDA),Open GL to construct the 3D pointcloud model,and the model is realistic. To resolve the problem of traditional point cloud display which requires high quality database,this paper proposes a dynamic linear recovering method to recover the holes which occur in 3D pointcloud model if it is based on low quality 3D data,and then use GPU to accelerate this process.This pointcloud data is inputed from the 3D face database and scanned progressively for holes occuring in the pointcloud,and the linear interpolating method which is consistent with human facial intrinsic feature to recover the holes is also used. The experiments show that the display effect could be far better than that the traditional point cloud display by our method,and the display efficiency can be greatly improved by using GPU acceleration.
引文
[1]PARKE F I.A parametric model of human faces[D].Salt Lake City:Ph D thesis,University of Utah,1974.
[2]PARKE F I,WATERS K.Computer facial animation[M].Massachusetts:AKPeters,Wellesley,1996.
[3]LEE Y C,TERZOPOULOS D,WATERS K.Realistic modeling for facial animation[C].Conference on Computer Graphics&Interactive Techniques,2010:55-62.
[4]LEE Y C,TERZOPOULOS D,WATERS K.Constructing physics-based facial models of individuals[C].Proceedings of Graphics Interface,1993:1-8.
[5]DECARLO D,METAXAS D,STONE M.An anthropometric face model using variational techniques[C].Computer Graphics,Annual Conference Series,1998:67-74.
[6]BLANZ V,VETTER T.A morphable model for the synthesis of 3D faces[C].Computer Graphics Proceedings Siggraph,1999.
[7]ZHANG ZH Y,LIU Z CH,ADLER D.Robust and rapid generation of animated faces from video images:a model-based modeling approach[J].2001.
[8]詹曙,张芝华,叶长明,等.三维人脸深度图的流形学习-LOGMAP识别方法[J].电子测量与仪器学报,2012,26(2):138-143.ZHAN SH,ZHANG ZH H,YE CH M,et al.3D facial depth map recognition based on manifold learningLOGMAP algorithm[J].Journal of Electronic Measurement and Instrument,2012,26(2):138-143.
[9]朱冰莲,王大为,李明均,等.针对表情变化的三维人脸识别系统研究[J].仪器仪表学报,2014,35(2):299-304.ZHU B L,WANG D W,LI M J,et al.Study on 3D face recognition system against expression change[J].Chinese Journal of Scientific Instrument,2014,35(2):299-304.
[10]闫勇刚,邓小玲,姚丽峰,等.基于主成分分析的大型管道三维表面重建[J].仪器仪表学报,2014,35(11):2567-2573.YAN Y G,DENG X L,YAO L F,et al.Threedimensional surface reconstruction of large-scale pipe combining principle component analysis[J].Chinese Journal of Scientific Instrument,2014,35(11):2567-2573.
[11]LINSEN L.Point cloud representation[R].Faculty of Computer Science,University of Karlsruhe,2001.
[12]ALEXA M,BEHR J,COHEN-OR D,et al.Pointset surfaces[C].Proceedings of IEEE Visualization,2001:21-28.
[13]陈飞舟,陈志杨,丁展,等.基于径向基函数的残缺点云数据修复[J].计算机辅助设计与图形学学报,2006,18(9):1414-1419.CHEN F ZH,CHEN ZH Y,DING ZH,et al.Filling holes in pointcloud with radial function[J].Journal of Computer-aided Design&Computer Graphics,2006,18(9):1414-1419.
[14]WANG J N,OLIVEIRAM M.A hole-filling strategy for reconstruction of smooth surfaces in range images[C].Proceedings of XVI Brazilian Symposium on Computer Graphics and Image Processing,2003:11.
[15]杨蕾.一种高效的支持向量回归的三维点云修补算法[J].计算机应用研究,2009,26(10):3945-3947.YANG L.Efficient pointcloud reconstruction algorithm based on support vector regression[J].Application Research of Computers,2009,26(10):3945-3947.
[2]PARKE F I,WATERS K.Computer facial animation[M].Massachusetts:AKPeters,Wellesley,1996.
[3]LEE Y C,TERZOPOULOS D,WATERS K.Realistic modeling for facial animation[C].Conference on Computer Graphics&Interactive Techniques,2010:55-62.
[4]LEE Y C,TERZOPOULOS D,WATERS K.Constructing physics-based facial models of individuals[C].Proceedings of Graphics Interface,1993:1-8.
[5]DECARLO D,METAXAS D,STONE M.An anthropometric face model using variational techniques[C].Computer Graphics,Annual Conference Series,1998:67-74.
[6]BLANZ V,VETTER T.A morphable model for the synthesis of 3D faces[C].Computer Graphics Proceedings Siggraph,1999.
[7]ZHANG ZH Y,LIU Z CH,ADLER D.Robust and rapid generation of animated faces from video images:a model-based modeling approach[J].2001.
[8]詹曙,张芝华,叶长明,等.三维人脸深度图的流形学习-LOGMAP识别方法[J].电子测量与仪器学报,2012,26(2):138-143.ZHAN SH,ZHANG ZH H,YE CH M,et al.3D facial depth map recognition based on manifold learningLOGMAP algorithm[J].Journal of Electronic Measurement and Instrument,2012,26(2):138-143.
[9]朱冰莲,王大为,李明均,等.针对表情变化的三维人脸识别系统研究[J].仪器仪表学报,2014,35(2):299-304.ZHU B L,WANG D W,LI M J,et al.Study on 3D face recognition system against expression change[J].Chinese Journal of Scientific Instrument,2014,35(2):299-304.
[10]闫勇刚,邓小玲,姚丽峰,等.基于主成分分析的大型管道三维表面重建[J].仪器仪表学报,2014,35(11):2567-2573.YAN Y G,DENG X L,YAO L F,et al.Threedimensional surface reconstruction of large-scale pipe combining principle component analysis[J].Chinese Journal of Scientific Instrument,2014,35(11):2567-2573.
[11]LINSEN L.Point cloud representation[R].Faculty of Computer Science,University of Karlsruhe,2001.
[12]ALEXA M,BEHR J,COHEN-OR D,et al.Pointset surfaces[C].Proceedings of IEEE Visualization,2001:21-28.
[13]陈飞舟,陈志杨,丁展,等.基于径向基函数的残缺点云数据修复[J].计算机辅助设计与图形学学报,2006,18(9):1414-1419.CHEN F ZH,CHEN ZH Y,DING ZH,et al.Filling holes in pointcloud with radial function[J].Journal of Computer-aided Design&Computer Graphics,2006,18(9):1414-1419.
[14]WANG J N,OLIVEIRAM M.A hole-filling strategy for reconstruction of smooth surfaces in range images[C].Proceedings of XVI Brazilian Symposium on Computer Graphics and Image Processing,2003:11.
[15]杨蕾.一种高效的支持向量回归的三维点云修补算法[J].计算机应用研究,2009,26(10):3945-3947.YANG L.Efficient pointcloud reconstruction algorithm based on support vector regression[J].Application Research of Computers,2009,26(10):3945-3947.