单Kinect+回转台的全视角三维重建

被引:11
作者
李健 [1 ]
李丰 [1 ]
何斌 [2 ]
杜希瑞 [1 ]
机构
[1] 陕西科技大学电气与信息工程学院
[2] 同济大学电气与信息工程学院
关键词
全视角; 三维重建; 转台; Kinect;
D O I
10.16337/j.1004-9037.2019.02.002
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
为了解决当前全视角三维扫描系统价格昂贵操作复杂的问题,提出利用1台Kinect和1个回转台来构建全视角三维模型的方法。研究涉及点云预处理、点云配准、全局误差修正以及色差修正等技术。首先使用Kinect采集数据并预处理,结合转台约束利用图像几何特征进行粗配准,随后使用迭代最近点(Iterative closest point,ICP)算法实现点云的精确配准。对于累积误差导致的闭环问题以及不同角度拍摄引起的色差问题,通过全局误差修正与色差修正算法处理,提升重建结果的精度。实验结果表明:该方法可以实现三维物体的全视角重建,并在精度上优于微软的KinectFusion方法。
引用
收藏
页码:205 / 213
页数:9
相关论文
共 17 条
[1]   Screened Poisson Surface Reconstruction [J].
Kazhdan, Michael ;
Hoppe, Hugues .
ACM TRANSACTIONS ON GRAPHICS, 2013, 32 (03)
[2]  
RGB-D mapping: Using Kinect-style depth cameras for dense 3D modeling of indoor environments. Peter Henry,Michael Krainin,Evan Herbst. INTERNATIONAL JOURNAL OF ROBOTICS RESEARCH . 2012
[3]  
Enhanced computer vision with Microsoft Kinect Sensor:a review. Han J G,Shao L,Xu D,et al. IEEE Transactions on Systems Man and Cybernetics . 2013
[4]  
A Method for Registration of 3-D Shapes. Paul J. Besl,Neil D. McKay. IEEE Transactions on Pattern Analysis and Machine Intelligence . 1992
[5]  
Multiview registration of 3D scenes by minimizing error between coordinate frames. Sharp, Gregory C.,Lee, Sang W.,Wehe, David K. IEEE Transactions on Pattern Analysis and Machine Intelligence . 2004
[6]  
Two algorithms for constructing a Delaunay triangulation[J] . D. T. Lee,B. J. Schachter. &nbspInternational Journal of Computer & Information Sciences . 1980 (3)
[7]  
Surf:Speeded up robust features. Bay H,Tuytelaars T,Van Gool L. Computer Vision-ECCV 2006 . 2006
[8]  
KinectFusion:Real-Time Dense Surface Mapping and Tracking. Newcombe R A,Izadi S,Hilliges O,et al. 2011 10th IEEE International Symposium on Mixed and Augmented Reality . 2011
[9]  
Kinect fusion:real-time 3D reconstruction and interaction using a moving depth camera. Izadi S,Kim D,Hilliges O,et al. 24th Annual ACM Symposium on User Interface Software and Technology . 2011
[10]  
Estimating the Laplace-Beltrami Operator by Restricting 3D Functions. Ming Chuang,Linjie Luo,Benedict J. Brown,Szymon Rusinkiewicz,Michael Kazhdan. Computer Graphics . 2009