|
|
Long-range 3D reconstruction based on wide-baseline stereo vision |
WANG Wei-qiang, XU Jin, DU Xin, LIU Ji-lin |
Institute of Information and Communication Engineering, Zhejiang University,
Zhejiang Provincial Key Laboratory of Information Network Technology, Hangzhou 310027,China |
|
|
Abstract A self-calibration based wide-baseline stereo vision technique was proposed for remote scene’s 3D reconstruction in vision based navigation of lunar rover. Correspondences between stereo rig were obtained by extracting and matching of scaleinvariant feature transform (SIFT) features. Then with the epipolar geometrical description of rectifying collineation, the leastsquares solution to Sampson error was achieved by Levenberg-Marquardt(LM) algorithm, which obtained the camera parameter and the rectified stereo rig. Following, the widebaseline dense matching was completed using seeds’ disparity expansion algorithm. Finally, target’s 3D reconstruction was achieved on the calculated disparity. This technique is demonstrated to have good performance on a number of remote natural mountain scenes and solve the problems of field calibration, change in illumination, perspective deformations and occlusions in widebaseline stereo vision.
|
Published: 16 July 2010
|
|
基于宽基线立体视觉的远距离三维重建
针对月球车视觉导航中远距离场景的三维重建问题,提出一种基于相机自标定的宽基线立体视觉方法.该方法首先提取立体图对的尺寸不变性(SIFT)特征并进行匹配,得到特征点对应关系;再使用外极线几何约束描述立体图像校正的直射变换,对其Sampson误差使用LevenbergMarquardt(LM)算法求得最小二乘最优解,完成相机参数估计和图像校正;对校正后的图像使用种子像素视差扩张算法进行宽基线立体匹配,根据匹配得到的视差完成目标场景三维重建.实验表明:该方法解决了月球车相机的现场标定难题,并能够解决宽基线立体视觉面临的光照差异、透视畸变、遮挡等问题,对远距离山脉等自然场景的三维重建效果良好,重建精度较高.
|
|
[1] PRITCHETT P, ZISSERMAN A. Wide baseline stereo matching [C]∥ Proceedings of the 6th International Conference on Computer Vision. Bombay, India: IEEE Computer Society, 1998: 754759.
[2] TUYTELAARS T, VAN GOOL L. Wide baseline stereo matching based on local, affinely invariant regions [C]∥ Proceedings of the 11th British Machine Vision Conference. Bristol, UK: ILES Central, 2000: 412425.
[3] TELL D, CADSSON S. Wide baseline point matching using invariants computed from intensity profiles [C]∥ Proceedings of the 6th European Conference on Computer Vision. Dublin, Ireland: Springer, 2000.
[4] LOWE D G. Distinctive image features from scaleinvariant key points [J]. International Journal of Computer Vision, 2004: 60(2):91110.
[5] IRSARA L, FUSIELLO A. Quasieuclidean uncalibrated epipolar rectification [C]∥ International Conference on Pattern Recognition (ICPR). Tampa, USA: \ [s.n.\], 2008: 14.
[6] HEYDEN A, POLLEFEYS M. Multiple view geometry [M]. NJ, USA: Prentice Hall, 2005.
[7] 沈晔湖,刘济林.利用立体图对的三维人脸模型重建算法[J].计算机辅助设计与图形学学报,2006,12(18): 19041909.
SHEN Yehu, LIU Jilin. From stereo pair to 3D face model [J]. Journal of ComputerAided Design & Computer Graphics, 2006, 12(18): 19041909.
[8] NovAtel产品类型大全[EB/OL].[20080823].http:∥www.novatel.com.cn/products/NovAtel%20product.pdf.
[9] Google Earth 4.3 [CP/OL]. [20080820]. http:∥earth.google.com/downloadearth.html. |
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
|
Shared |
|
|
|
|
|
Discussed |
|
|
|
|