基于光线的全局优化多视图三维重建方法

doi:10.3969/j.issn.1000-3428.2013.11.053

摘要/Abstract

摘要： 利用光线跟踪原理，提出一种全局优化的多视图三维重建方法。根据图像轮廓得到物体的包围盒，采用体素离散物体所在的几何空间。从相机中心向图像上每个像素发射一条光线，为确定光线达到的体素，使用归一化互相关(NCC)值度量光线-体素的一致性，并估计采样空间中面片的法向信息，以提高NCC值的可信度。设计基于因子图的全局优化模型得到物体体素，针对光线因子的特殊性，设计一种高效的置信度传播算法，使重建方法的时间复杂度从指数阶降为线性阶。实验结果表明，与基于马尔可夫场的重建方法相比，该方法的鲁棒性较好，可提高重建模型的准确度和完整性。

关键词: 三维重建, 因子图, 可视壳, 多视图立体匹配, 归一化互相关

Abstract: This paper presents a multi-view 3D reconstruction method based on rays. It generates the bounding volume for the object in the images using silhouettes, and discretizes space of the object into voxels. For each pixel in the image, a ray is generated from the center of the camera. To determine the voxels for the rays, Normalized Cross-correlation(NCC) value is used to measure the consistence between the rays and the voxels, where the patch normal is estimated to improve the confidence of the corresponding NCC value. A global optimization model based on factor graph is designed to extract the voxels belonging to the reconstructed object. In addition, an efficient belief propagation algorithm is proposed based on the characteristics of ray factor, which reduces the computational complexity from an exponential one to a linear one. Experimental results show that the proposed method is more robust than the previous reconstruction method based on Markov Random Field(MRF), and achieves better reconstruction result in terms of both accuracy and completeness.

Key words: 3D reconstruction, factor graph, visual hull, multi-view stereo matching, Normalized Cross Correlation(NCC)

中图分类号:

TP18

陈坤，刘新国. 基于光线的全局优化多视图三维重建方法[J]. 计算机工程, doi: 10.3969/j.issn.1000-3428.2013.11.053.

CHEN Kun, LIU Xin-guo. Global Optimized Multi-view 3D Reconstruction Method Based on Rays[J]. Computer Engineering, doi: 10.3969/j.issn.1000-3428.2013.11.053.

参考文献

参考文献 [1] 段春梅. 基于多视图的三维模型重建方法研究[D]. 济南: 山东大学, 2009. [2] Seitz S, Curless B, Diebel J, et al. A Comparison and Evalua- tion of Multi-view Stereo Reconstruction Algorithms[C]//Proc. of CVPR’06. New York, USA: IEEE Computer Society, 2006. [3] Vogiatzis G, Torr P H S, Cipolla R. Multi-view Stereo via Volumetric Graph-cuts[C]//Proc. of CVPR’05. San Diego, USA: IEEE Computer Society, 2005. [4] Boykov Y, Kolmogorov V. Computing Geodesics and Minimal Surfaces via Graph Cuts[C]//Proc. of CVPR’03. Madison, USA: IEEE Computer Society, 2003. [5] Boykov Y, Veksler O, Zabih R. Fast Approximate Energy Minimization via Graph Cuts[J]. PAMI, 2001, 23(11): 1222- 1239. [6] Starck J, Hilton A, Miller G. Volumetric Stereo with Silhouette and Feature Constraints[C]//Proc. of British Machine Vision Conference. Edinburgh, UK: [s. n.], 2006. [7] Goesele S, Curless B, Seitz S. Multi-view Stereo Revisited[C]// Proc. of CVPR’06. New York, USA: IEEE Computer Society, 2006. [8] Liu Shubao, Cooper D B. Ray Markov Random Fields for Image-based 3D Modeling: Model and Efficient Inference[C]// Proc. of CVPR’10. San Francisco, USA: IEEE Computer Society, 2010. [9] Liu Shubao, Cooper D B. A Complete Statistical Inverse Ray Tracing Approach to Multi-view Stereo[C]//Proc. of CVPR’11. Colorado Springs, USA: IEEE Computer Society, 2011. [10] Kutulakos K, Seitz S. A Theory of Shape by Space Carving[J]. International Journal of Computer Vision, 2000, 38(3): 199- 218. [11] Hernandez C, Schmitt F. Silhouette and Stereo Fusion for 3D Object Modeling[J]. Computer Vision and Image Under- standing, 2004, 96(3): 367-392. [12] Zaharescu A, Boyer E, Horaud R P. TransforMesh: A Topology-adaptive Mesh-based Approach to Surface Evolution[C]//Proc. of ACCV’07. Tokyo, Japan: [s. n.], 2007. [13] Bradley D, Boubekeur T, Berlin T. Accurate Multi-view Reconstruction Using Robust Binocular Stereo and Surface Meshing[C]//Proc. of CVPR’08. [S. l.]: IEEE Computer Society, 2008. [14] Furukawa Y, Ponce J. Accurate, Dense and Robust Multi-view Stereopsis[C]//Proc. of CVPR’07. Minneapolis, USA: IEEE Computer Society, 2007. [15] Habbecke M, Kobbelt L. A Surface-growing Approach to Multi-view Stereo Reconstruction[C]//Proc. of CVPR’07. Minneapolis, USA: IEEE Computer Society, 2007. [16] Jancosek M, Pajdla T. Segmentation Based Multi-view Stereo[C]//Proc. of Computer Vision Winter Workshop. [S. l.]: IEEE Press, 2009. [17] Wu Tai-Pang, Yeung Sai-Kit, Jia Jiaya, et al. Quasi-dense 3D Reconstruction Using Tensor-based Multi-view Stereo[C]// Proc. of CVPR’10. San Francisco, USA: IEEE Computer Society, 2010. [18] Esteban C H. Stereo and Silhouette Fusion for 3D Object Modeling from Uncalibrated Images Under Circular Motion[D]. Pairs, France: Nationale Sup?ieure des Télécom- munications, 2004. [19] Kschischang F R, Frey B J, Loeliger H A. Factor Graphs and the Sum-product Algorithm[J]. IEEE Transactions on Information Theory, 2001, 47(2): 498-519. 编辑金胡考