|
Improved Stereo Matching With Boosting Method
[Full Text]
AUTHOR(S)
Shiny B, Dr. Deepa J.
KEYWORDS
Index Terms: Stereo matching, occlusion handling, initial processing, classification, completion stage, median filtering, performance evaluation
ABSTRACT
Abstract: This paper presents an approach based on classification for improving the accuracy of stereo matching methods. We propose this method for occlusion handling. This work employs classification of pixels for finding the erroneous disparity values. Due to the wide applications of disparity map in 3D television, medical imaging, etc, the accuracy of disparity map has high significance. An initial disparity map is obtained using local or global stereo matching methods from the input stereo image pair. The various features for classification are computed from the input stereo image pair and the obtained disparity map. Then the computed feature vector is used for classification of pixels by using GentleBoost as the classification method. The erroneous disparity values in the disparity map found by classification are corrected through a completion stage or filling stage. A performance evaluation of stereo matching using AdaBoostM1, RUSBoost, Neural networks and GentleBoost is performed.
REFERENCES
[1] M.Baydoun, M.A.Al-Alaoui,"Enhancing stereo matching with classification", IEEE Access, Digital Object Identifier 10.1109/ACCESS.2014.2322101.
[2] S. Birchfield and C. Tomasi, ``A pixel dissimilarity measure that is insensitive to image sampling,'' IEEE Trans. Pattern Anal. Mach. Intell., vol. 20,no. 4, pp. 401-406, Apr. 1998.
[3] T. Kanade and M. Okutomi, ``A stereo matching algorithm with an adaptive window: Theory and experiment,'' IEEE Trans. Pattern Anal. Mach. Intell., vol. 16, no. 9, pp. 920-932, Sep. 1994.
[4] K.-J. Yoon and I.-S. Kweon, ``Locally adaptive support-weight approach for visual correspondence search,'' in Proc. IEEE Comput. Soc. Conf. CVPR, vol. 2. Jun. 2005, pp. 924-931.
[5] V. Vineet and P. J. Narayanan, ``CUDA cuts: Fast graph cuts on the GPU,''in Proc. IEEE Comput. Soc. Conf. CVPRW Jun. 2008, pp. 1-8.
[6] M.-H. Ju and H.-B. Kang, ``A new method for stereo matching using pixel cooperative optimization,'' in Proc. 16th IEEE ICIP, Nov. 2009, pp. 2105-2108.
[7] Z.-F. Wang and Z.-G. Zheng, ``A region based stereo matching algorithm using cooperative optimization,'' in Proc. IEEE Conf. CVPR, Jun. 2008,pp. 1-8.
[8] J. Y. Goulermas, P. Liatsis, and T. Fernando, ``A constrained nonlinear energy minimization framework for the regularization of the stereo correspondence problem,'' IEEE Trans. Circuits Syst. Video Technol., vol. 15,no. 4, pp. 550-565, Apr. 2005.
[9] E. Park and K. Wohn, ``Stereo and motion correspondences using nonlinear optimization method,'' Comput. Vis. Image Understand., vol. 101, no. 3, pp. 194-203, 2006.
[10] W. Kim, J. Park, and K. Lee, ``Stereo matching using population-based MCMC,'' Int. J. Comput. Vis., vol. 83, no. 2, pp. 195-209, 2009.
[11] P. F. Felzenszwalb and D. P. Huttenlocher, ``Ef cient belief propagation for early vision,'' Int. J. Comput. Vis., vol. 70, no. 1, pp. 41-54, 2006.
[12] Q. Yang, L. Wang, and N. Ahuja, ``A constant-space belief propagation algorithm for stereo matching,'' in Proc. IEEE Conf. CVPR, Jun. 2010, pp.1458-1465.
[13] I. Ernst and H. Hirschmüller, ``Mutual information based semi-global stereo matching on the GPU,'' in Proc. 4th Int. Symp. Adv. Vis. Comput., 2008, pp. 228-239.
[14] K.-J. Yoon and I. S. Kweon, ``Stereo matching with the distinctive similarity measure,'' in Proc. IEEE ICCV, Oct. 2007, pp. 1-7.
[15] J. Gibson and O. Marques, ``Stereo depth with a uni ed architecture GPU,'' in Proc. IEEE Comput. Soc. Conf. CVPRW, Jun. 2008, pp. 1-6.
[16] J. Sun, Y. Li, S. B. Kang, and H.-Y. Shum, ``Symmetric stereo matching for occlusion handling,'' in Proc. IEEE Comput. Soc. Conf. CVPR, vol. 2. Jun. 2005, pp. 399-406.
[17] A. F. Bobick and S. S. Intille, ``Large occlusion stereo,'' Int. J. Comput. Vis., vol. 33, no. 3, pp. 181-200, 1999.
[18] C. L. Zitnick and T. Kanade, ``A cooperative algorithm for stereo matching and occlusion detection,'' IEEE Trans. Pattern Anal. Mach. Intell., vol. 22, no. 7, pp. 675-684, Jul. 2000.
[19] M. Baydoun and M. A. Al-Alaoui, ``Enhancing stereo matching with varying illumination through histogram information and normalized cross correlation,'' in Proc. 20th IWSSIP, Jul. 2013, pp. 5-9.
[20] A. Hosni, C. Rhemann, M. Bleyer, C. Rother, and M. Gelautz, ``Fast cost-volume filtering for visual correspondence and beyond,'' IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 2, pp. 504-511, Feb. 2013.
[21] A. Hosni, M. Bleyer, C. Rhemann, M. Gelautz, and C. Rother,``Real-time local stereo matching using guided image filtering,'' in Proc. IEEE Int. Conf. ICME, Jul. 2011, pp. 1-6.
[22] W.-H. Lee, Y. Kim, and J. B. Ra, ``Efficient stereo matching based on a new confidence metric,'' in Proc. 20th EUSIPCO, 2012, pp. 1139-1143.
[23] D. Scharstein and R. Szeliski, ``High-accuracy stereo depth maps using structured light,'' in Proc. IEEE Comput. Soc. Conf. CVPR, vol. 1. Jun. 2003, pp. I-195-I-202.
|
