Evaluation of Stereo Matching Algorithms and Dynamic Programming for 3D Triangulation

A good result of triangulation or known as Three-Dimensional (3D) is depending on the smoothness of the disparity depth map that obtained from the stereo matching algorithms. The smoother the disparity depth map, the better results of triangulation can be achieved. This paper presents the evaluation of the existing stereo matching algorithms in the aspects of the speed of computational on depth map obtained. The stereo matching algorithms that we applied for experimental purpose are basic block matching, sub-pixel accuracy and dynamic programming. The dataset of stereo images that used for the experimental purpose are obtained from Middlebury Stereo Datasets. This research is to provide an idea on choosing the better stereo matching algorithms to work on the disparity depth map for the purpose of 3D triangulation applications, as the good result of 3D triangulation is depending on how smooth is the disparity depth map can be obtained.

[1]  Margrit Gelautz,et al.  A layered stereo matching algorithm using image segmentation and global visibility constraints , 2005 .

[2]  Richard Szeliski,et al.  A Taxonomy and Evaluation of Dense Two-Frame Stereo Correspondence Algorithms , 2001, International Journal of Computer Vision.

[3]  Andrew Blake,et al.  Visual Reconstruction , 1987, Deep Learning for EEG-Based Brain–Computer Interfaces.

[4]  Ying Wang,et al.  Efficient and accurate subpixel path based stereo matching , 2008, 2008 19th International Conference on Pattern Recognition.

[5]  Luigi di Stefano,et al.  An efficient algorithm for exhaustive template matching based on normalized cross correlation , 2003, 12th International Conference on Image Analysis and Processing, 2003.Proceedings..

[6]  Anders Olofsson,et al.  Modern Stereo Correspondence Algorithms : Investigation and Evaluation , 2010 .

[7]  Daniel P. Huttenlocher,et al.  Efficient Belief Propagation for Early Vision , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[8]  Federico Tombari,et al.  A 3D reconstruction system based on improved spacetime stereo , 2010, 2010 11th International Conference on Control Automation Robotics & Vision.

[9]  Carlo Tomasi,et al.  Depth Discontinuities by Pixel-to-Pixel Stereo , 1999, International Journal of Computer Vision.

[10]  Olga Veksler,et al.  Stereo correspondence by dynamic programming on a tree , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[11]  Yi-Ping Hung,et al.  Fast block matching algorithm based on the winner-update strategy , 2001, IEEE Trans. Image Process..

[12]  Donald Geman,et al.  Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images , 1984, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Federico Girosi,et al.  Parallel and Deterministic Algorithms from MRFs: Surface Reconstruction , 1991, IEEE Trans. Pattern Anal. Mach. Intell..

[14]  Vladimir Kolmogorov,et al.  Multi-camera Scene Reconstruction via Graph Cuts , 2002, ECCV.

[15]  Olle Sundström,et al.  A generic dynamic programming Matlab function , 2009, 2009 IEEE Control Applications, (CCA) & Intelligent Control, (ISIC).