[1]贾贝贝,阮秋琦.双目立体视觉的三维人脸重建方法[J].智能系统学报,2009,4(06):513-520.[doi:10.3969/j.issn.1673-4785.2009.06.008]
 JIA Bei-bei,RUAN Qiu-qi.3D face reconstruction using binocular stereo vision[J].CAAI Transactions on Intelligent Systems,2009,4(06):513-520.[doi:10.3969/j.issn.1673-4785.2009.06.008]
点击复制

双目立体视觉的三维人脸重建方法(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第4卷
期数:
2009年06期
页码:
513-520
栏目:
出版日期:
2009-12-25

文章信息/Info

Title:
3D face reconstruction using binocular stereo vision
文章编号:
1673-4785(2009)06-0513-08
作者:
贾贝贝阮秋琦
北京交通大学 计算机与信息技术学院,北京 100044
Author(s):
JIA Bei-bei RUAN Qiu-qi
School of Computer Science and Information Technology, Beijing Jiaotong University, Beijing 100044, China
关键词:
三维人脸模型双目立体视觉摄像机标定极线几何立体匹配区域生长三角剖分
Keywords:
3D face modelbinocular stereo visioncamera calibrationepipolar geometrystereo matchingregion growingdelaunay triangulation
分类号:
TP391
DOI:
10.3969/j.issn.1673-4785.2009.06.008
文献标志码:
A
摘要:
创建逼真的三维人脸模型始终是一个极具挑战性的课题.随着三维人脸模型在虚拟现实、视频监控、三维动画、人脸识别等领域的广泛应用,三维人脸重建成为计算机图像学和计算机视觉领域的一个研究热点.针对这一问题,提出一种基于双目立体视觉的三维人脸重建方法,重建过程中无需三维激光扫描仪和通用人脸模型.首先利用标定的2台摄像机获取人脸正面图像对,通过图像校正使图像对的极线对齐并且补偿摄像机镜头的畸变;在立体匹配方面,选择具有准确可靠视差的人脸边缘特征点作为种子像素,以种子像素的视差作为区域生长的视差,在外极线约束、单调性约束以及对应匹配的边缘特征点的约束下,进行水平扫描线上的区域生长,从而得到整个人脸区域的视差图,提高了对应点匹配的速度和准确度;最后,根据摄像机标定结果和立体匹配生成的视差图计算人脸空间散乱点的三维坐标,对人脸的三维点云进行三角剖分、网格细分和光顺处理.实验结果表明,该方法能够生成光滑、逼真的三维人脸模型,证明了该算法的有效性.
Abstract:
Generation of lifelike 3D human faces is a challenging task. Recent increases in the use of 3D face models in virtual reality, video surveillance, 3D animation, and face recognition have led to 3D face reconstruction becoming a research hotspot. The authors proposed a 3D face reconstruction method based on binocular stereo vision theory. After capturing the front view of a face with two calibrated cameras, the captured pair of stereo images were rectified to align their epipolar lines and compensate for image distortions. To obtain maps with accurate matching and dense disparity, a stereo matching algorithm based on region growing was developed. An edge feature point with reliable disparity was selected as a seed point. Then region growing was performed along horizontal scanlines under multiconstraints. With camera calibration and disparity mapping, 3D coordinates of corresponding points were calculated. Then the face model was reconstructed with a series of methods, such as Delaunay triangulation, mesh subdivision and smoothing. Experimental results showed that the method can generate a smooth and lifelike 3D face model.

参考文献/References:

[1]PARKE F I. Computer generated animation of faces[C]//Proceedings of ACM Annual Conference. Boston, United States, 1972:451-457.
[2]LEE Y C, TERZOPOULOS D, WATERS K. Realistic face modeling for animation[C]//SIGGRAPH Proceeding. Los Angeles, USA, 1995:55-62.
[3]BLANZ V, VETTER T. A morphable model for the synthesis of 3D faces[C]//SIGGRAPH Proceedings. Orlando, FL, USA, 1999:187-194.
[4]ANSARI A, ABDELMOTTALEB M. 3D face modeling using two views and a generic face model with application to 3D face recognition[C]//Proceedings of the IEEE Conference on Advanced Video and Signal Based Surveillance. Miami, USA, 2003:37-44.
[5]LIU Z, ZHANG Z, JACOBS C, et al. Rapid modeling of animated faces from video M SR-TR-2000-11[R]. 2000.
[6]王宣银, 潘    峰, 向桂山,等. 基于Snake模型的特定人脸三维重建方法[J]. 机械工业学报, 2007, 43(7):168-173.
WANG Xuanyin, PAN Feng, XIANG Guishan, et al. Specific face 3D reconstruction based on Snake model[J]. Chinese Journal of Mechanical Engineering, 2007, 43(7):168-173.
[7]CHEN Q, MEDIONI G. Building 3-D human face models from two photographs[J]. The Journal of VLSI Signal Processing, 2001, 27(2):127-140.
[8]张广军. 机器视觉[M]. 北京: 科学出版社, 2005:58-84.
[9]ZHANG Zhengyou. A flexible new technique for camera calibration[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(11):1330-1334.
[10]MYRON Z, DARIUS BURSCHKA, GREGORY D. Advances in computational stereo[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(8):993-1008.
[11]阮秋琦. 数字图像处理学[M]. 2版. 北京: 电子工业出版社, 2007:382-408.
[12]SHEWCHUK J R. Delaunay refinement algorithms for triangular mesh generation[J]. Computational Geometry— Theory And applications, 2002,22(113):21-74.
[13]张永春, 达飞鹏, 宋文忠. 三维散乱点集的曲面三角剖分[J]. 中国图象图形学报,2003, 8(12):1379-1388.
ZHANG Yongchun, DA Feipeng, SONG Wenzhong. Surface triangulations based on arbitrary pointsets[J]. Journal of Image and Graphics,2003, 8(12):1379-1388.
[14]KOBBELT L. A variational approach subdivision[J]. Computer Aided Geometric Design, 1996,13(8):743-761.
[15]FIELD D A. Laplacian smoothing and Delaunay triangulations[J]. Communications in Applied Numerical Methods, 1988,(4):709-712.

相似文献/References:

[1]张凤静,周建江,夏伟杰.基于双目立体视觉的汽车安全车距测量方法[J].智能系统学报,2011,6(01):79.
 ZHANG Fengjing,ZHOU Jianjiang,XIA Weijie.Research of vehicle security spacing measurements based on binocular stereovision[J].CAAI Transactions on Intelligent Systems,2011,6(06):79.
[2]尚倩,阮秋琦,李小利.双目立体视觉的目标识别与定位[J].智能系统学报,2011,6(04):303.
 SHANG Qian,RUAN Qiuqi,LI Xiaoli.Target recognition and location based on binocular stereo vision[J].CAAI Transactions on Intelligent Systems,2011,6(06):303.

备注/Memo

备注/Memo:
收稿日期:2009-04-15.
基金项目:国家自然科学基金资助项目(60672062);高等学校博士学科点专项科研基金资助项目(200800040008).
作者简介:
贾贝贝,女,1985年生,硕士研究生,主要研究方向为计算机视觉、数字图像处理、信号与信息处理.
阮秋琦,男,1944年生,教授,博士生导师,北京交通大学信息科学研究所所长,国务院学位委员会学科评议组成员,IEEE高级会员. 主要研究方向为图像处理、计算机视觉、多媒体信息处理、虚拟现实、计算机支持的协同工作. 曾获国家教委科技进步二等奖、铁道部科技进步二等奖和三等奖等,被评为国家级有突出贡献的中青年专家,获茅以升科技奖、詹天佑科技人才奖并享受国家政府津贴的荣誉. 发表学术论文150余篇,出版著作3部,获国家专利1项.
更新日期/Last Update: 2010-02-22