Abstract
This paper proposes a stochastic approach to estimate the disparity field combined with line field. In the maximum a posteriori (MAP) method based on Markov random field (MRF) model, it is important to optimize and converge the Gibbs potential function corresponding to the perturbed disparity field. The proposed optimization method, stochastic diffusion, takes advantage of the probabilistic distribution of the neighborhood fields to diffuse the Gibbs potential space iteratively. By using the neighborhood distribution in the non-random and non-deterministic diffusion, both the estimation accuracy and the convergence speed are improved. In the paper, the hierarchical stochastic diffusion is also applied to the disparity field. The hierarchical approach reduces the memory and computational load, and increases the convergence speed of the potential space. The paper also proposes an effective configuration of the neighborhood to be suitable for the hierarchical disparity structure. According to the experiments, the stochastic diffusion shows good estimation performance. The line field improves the estimation at the object boundary, and coincides with the object boundary with the useful contours. The stochastic diffusion is applicable to any kind of field estimation given the appropriate definition of the field and MRF models.
Similar content being viewed by others
References
Barnard, S.T. 1989. Stochastic stereo matching over scale. IJCV, 3:17–32.
Besag, J. 1974. Spatial interaction and the statistical analysis of lattice system. J. Royal Stat. Soc. B, (2):192–236.
Black, M. and Fleet, D. 2000. Probabilistic detection and tracking of motion boundaries. International Journal of Computer Vision, 38(3):231–245.
Boykov, Y., Veksler, O., and Zahih, R. 1999. Fast approximation energy minimization via graph cuts. In Proc. of CVPR-99.
Burt, P.J. and Adelson, E.H. 1983. The Laplacian pyramid as a compact image code. IEEE Trans. on Communications, COM-31(4): 532–540.
Chang, M., Tekalp, A.M., and Sezan, M.I. 1997. Simultaneous motion estimation and segmentation. IEEE Trans. Image Processing, 6(9):1326–1333.
Chang, N.L. and Zakhor, A. 1997. View generation for three-dimensional scenes from video sequences. IEEE Trans. Image Processing, 6(4):584–598.
Chellappa, R. and Jain, A. 1993. Markov Random Fields. Academic Press: San Mateo, CA.
Faugeras, O. 1993. Three-Dimensional Computer Vision. A Geometric Viewpoint. MIT Press: Cambridge, MA.
Geman, S. and Geman, D. 1984. Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images. IEEE Trans. on PAMI, PAMI-6(6):721–741.
Gonzalez, 1995. Digital Image Processing. Addison Wiely: Reading, MA.
Gordon, N.J., Salmond, D.J., and Smith, A.F.M. 1993. Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proceedings-F, 140(2):107–113.
Graffigne, C. et al. 1995. Hierarchical Markov random field models applied to image analysis: A review. In Proc. of VCIP of SPIE, SPIE 2568, pp. 2–17.
Haralick, R.M. and Shapiro, L.G. 1993. Computer and Robot Vision. Addison Wesley: Reading, MA, 1993.
Heitz, F. and Bouthemy, P. 1993. Multimodal estimation of discontinuous optical flow using Markov random field. IEEE Trans. on PAMI, 15(12):1217–1232.
Heitz, F., Perez, P., and Bouthemy, P. 1994. Multiscale minimization of global energy functions in some visual recovery problems. CVGIP: Image Understanding, 59(1):125–134.
Isard, M. and Blake, A. 1998. Condensation-conditional density propagation for visual tracking. International Journal of Computer Vision, 29(1):5–28.
Jebara, T., Azarbayejani, A., and Pentland, A. 1999. 3D structure from 2D motion. IEEE Signal Processing Magazine, pp. 66–84.
Kanade, T. and Okutomi, M. 1994. A stereo matching algorithm with an adaptive window: Theory and experiments. IEEE Trans. on PAMI, PAMI-16(9):920–932.
Kim, M., Choi, J. et al., and Ho, Y. 1999. A VOP generation tool: Automatic segmentation of moving objects in image sequences based on spatio-temporal information. IEEE Trans. on Circuits and Syst. for Video Tech., 9(8):1216–1226.
Konrad, J. and Dubois, E. 1992. Bayesian estimation of motion vector fields. IEEE Trans. on PAMI, 14:910–927.
Kreyzig, E. 1978. Introductory Functional Analysis with Applications. WIE Wiley: New York.
Lee, S.H., Park, J.-I., and Lee, C.W. 1998. A new stereo matching algorithm based on Baysian model. In Proc. of ICASSP-98, Seattle, USA, Vol. 5, pp. 2769–2772, May, 1998.
Lee, S.H., Park, J.-I., Inoue, S., and Lee, C.W. 1999. Disparity estimation based on Bayesian maximum a posteriori (MAP) algorithms. IEICE Trans. on Fund. of Elec., Commun., and Comp. Sci., E82-A(7):1367–1376.
Lee, S.H., Park, J.-I., and Lee, C.W. 2000. Correspondence and line field estimation using MAP-based probabilistic diffusion algorithm. In Proc. of ICIP-2000, pp. 2213–2216.
Lee, S.H., Kanatsugu, Y., and Park, J.-I. 2001. MAP-based simultaneous correspondence estimation and object segmentation using stochastic diffusion. In Proc. of Picture Coding Symposium, PCS-2001, Seoul, Korea, April 2001.
Li, Ze-Nian and Hu, Gongzhu. 1996. Analsis of disparity gradient based cooperative stereo. IEEE Trans. on Image Processing, 5(11):1493–1506.
Marroquin, J., Mitter, S., and Poggio, T. 1987. Probabilistic solution of ill-posed problems in computational vision. Journal of the American Statistical Association, 82(397):76–89.
Meier, T. and Ngan, K.N. 1999. Video segmentation for content-based coding. IEEE Trans. on Circuits and Syst. for Video Tech., 9(8):1190–1203.
Meyer, F. and Bouthemy, P. 1994. Region-based tracking using affine motion models in long image sequences. CVGIP: Image Understanding, 60(2):119–140.
MPEG-4 video verification model version 8.0. 1997. ISO/IEC JTC1/SC29/ WG11, Draft.
Motion Picture Exports Group (JTC1/SC29/WG11) and Experts Group on ATM Video Coding (ITU-T SG15), 1994. Generic coding of moving pictures and associated audio MPEG-2. Draft International Standard 13813, ISO/IEC.
Moellenhoff, M. and Maier, M.W. 1998. Transform coding of stereo image residuals. IEEE Trans. on Image Processing, 7(6):804–812.
Nadabar, S.G. and Jain, A.K. 1996. Parameter estimation in Markov random field contextual models using geometric models of objects. IEEE Trans. on PAMI, 18(3):326–329.
Redert, P., Hendricks, E., and Biemond, J. 1997. Synthesis of multi-viewpoint images at non-intermediate positions. In Proc. of ICASSP97, pp. 2749–2852.
Redert, A., Hendricks, E., and Biemond, J. 1999. Correspondence estimation in image pairs. IEEE Signal Processing Magazine, 16(3):29–46.
Scharstein, D. and Szeliski, R. 1998. Stereo matching with non-linear diffusion. International Jour. Comp. Vision, 28(2):155–174.
Scharstein, D. and Szeliski, R. 2002. A taxonomy and evaluation of dense two-frame stereo correspondence algorithm. IJCV.
Stiller, C. 1993. A statistical image model for motion estimation. In Proc. of ICASSP-93, V:193–196.
Stiller, C. 1997. Object-based estimation of dense motion field. IEEE Trans. on Image Proc., 6(2):234–250.
Szeliski, R. and Zabih, R. 1999. An experimental comparison of stereo algorithms. In Proc. of Vision Algorithms: Theory and Practice.
Tekalp, A.M. 1995. Digital Video Processing. Prentice Hall: Englewood Cliffs, NJ.
Torr, P.H.S., Szeliski, R., and Anandan, P. 1999. An integrated Bayesian approach to layer extraction from image sequence. In Proc. of ICCV.
Woo, W.T. and Ortega, A. 1996. Stereo image compression with disparity compensation using MRFmodel. In Proc. of SPIE VCIP-96, vol. 2727, Feb. 1996, pp. 28–41.
Zhang, J. 1992. The mean field theory in EM procedures for Markov random fields. IEEE Trans. on Signal Proc., 40(10):2570–2583.
Zhang, J. and Hanauer, G. 1995. The application of mean field theory to image motion estimation. IEEE Trans. on IP, 4(1):19–32.
Zitnick, C.L. and Kanade, T. 2000. A cooperative algorithm for stereo matching and occlusion detection. IEEE Trans. on PAMI, PAMI-22(7):675–684.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Lee, S.H., Kanatsugu, Y. & Park, JI. MAP-Based Stochastic Diffusion for Stereo Matching and Line Fields Estimation. International Journal of Computer Vision 47, 195–218 (2002). https://doi.org/10.1023/A:1014550009499
Issue Date:
DOI: https://doi.org/10.1023/A:1014550009499