This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Parallel and Deterministic Algorithms from MRFs: Surface Reconstruction
May 1991 (vol. 13 no. 5)
pp. 401-412

Deterministic approximations to Markov random field (MRF) models are derived. One of the models is shown to give in a natural way the graduated nonconvexity (GNC) algorithm proposed by A. Blake and A. Zisserman (1987). This model can be applied to smooth a field preserving its discontinuities. A class of more complex models is then proposed in order to deal with a variety of vision problems. All the theoretical results are obtained in the framework of statistical mechanics and mean field techniques. A parallel, iterative algorithm to solve the deterministic equations of the two models is presented, together with some experiments on synthetic and real images.

[1] M. Bertero, T. A. Poggio, and V. Torre, "Ill-posed problems in early vision,"Proc. IEEE, vol. 76, pp. 869-889, Aug. 1988.
[2] A. Blake and A. Zisserman,Visual Reconstruction. Cambridge, MA: MIT Press, 1987.
[3] J. F. Canny, "Finding lines and edges in images," Artificial Intell. Lab., Massachusetts Inst. Technol., Tech. Rep. TM-720, 1983.
[4] P. B. Chou and C. M. Brown, "Multimodal reconstruction and segmentation with Markov random fields and HCF optimization," inProc. Image Understanding Workshop. San Mateo, CA: Morgan Kaufmann, 1988, pp. 214-221.
[5] E. Gamble, D. Geiger, T. Poggio, and D. Weinshall, "Integration of vision modules and labeling of surface discontinuities,"IEEE Trans. Syst. Man, Cybern., Dec. 1989.
[6] E. B. Gamble and T. Poggio, "Visual integration and detection of discontinuities: The key role of intensity edges," Artificial Intell. Lab., Massachusetts Inst. Technol., AI Memo 970, Oct. 1987.
[7] D. Geiger and F. Girosi, "Parallel and deterministic algorithms for mrfs: surface reconstruction and integration," Artificial Intell. Lab., Massachusetts Inst. Technol., AI Memo 1114, May 1989.
[8] D. Geiger and R. M. Pereira, "On intelligent sparse data," inProc. IEEE 7th Israeli Symp. Artificial Intelligence and Computer Vision, Haifa, Israel, 1990.
[9] D. Geiger and T. Poggio, "An optimal scale for edge detection," inProc. IJCAI, Aug. 1987.
[10] D. Geiger and A. Yuille, "A common framework for image segmentation,"Int. J. Comput. Vision, 1990, to be published.
[11] S. Geman and D. Geman, "Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-6, pp. 721-741, 1984.
[12] F. Girosi, T. Poggio, and B. Caprile, "Extensions of a theory of networks for approximation and learning: Outliers and negative examples," Artificial Intell. Lab., Massachusetts Inst. Technol., Tech. Rep., 1989.
[13] J. Harris, C. Koch, and J. Luo, "A two-dimensional analog vlsi circuit for detecting discontinuities in early vision,"Science, pp. 1209-1211, 1990.
[14] S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, "Optimization by simulated annealing,"Science, vol. 220, pp. 219-227, 1983.
[15] C. Koch, J. Marroquin, and A. Yuille, "Analog 'neuronal' networks in early vision," inProc. Nat. Acad. Sci., vol. 83, pp. 4263-4267, 1987.
[16] A. Lumsdaine, J. Wyatt, and I. Elfadel, "Nonlinear analog networks for image smoothing and segmentation,"VLSI Signal Processing, 1990, to be published.
[17] J. L. Marroquin, "Deterministic Bayesian estimation of Markovian random fields with applications to computational vision," inProc. Int. Conf. Computer Vision, London, England, June 1987. Washington, DC: IEEE Press.
[18] J. L. Marroquin, S. Mitter, and T. Poggio, "Probabilistic solution of ill-posed problems in computational vision," inProc. Image Understanding Workshop, L. Baumann, Ed., McLean, VA, Aug. 1985, pp. 293-309.
[19] N. Metropolis, A. Rosenbluth, M. Rosenbluth, A. Teller, and E. Teller, "Equation of state calculations by fast computing machines,"J. Phys. Chem., vol. 21, p. 1087, 1953.
[20] D. Terzopoulos, "Multiresolution computation of visible-surface representation," Ph.D. dissertation, Massachusetts Inst. Technol., Cambridge, 1984.
[21] A. N. Tikhonov and V. Y. Arsenin,Solutions of Ill-Posed Problems. Washington, DC: Winston, 1977.
[22] G. Wahba, "Practical approximate solutions to linear operator equations when the data are noisy,"SIAM J. Numer. Anal., vol. 14, 1977.
[23] J. Zerubia and R. Chellappa, "Mean field annealing for edge detection and image restoration," inProc. European Signal Processing Conf., Barcelona, Spain, Sept. 1990.

Index Terms:
parallel algorithms; picture processing; Markov random field model; deterministic algorithms; surface reconstruction; statistical mechanics; mean field techniques; iterative algorithm; iterative methods; Markov processes; parallel algorithms; picture processing; statistical analysis
Citation:
D. Geiger, F. Girosi, "Parallel and Deterministic Algorithms from MRFs: Surface Reconstruction," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 13, no. 5, pp. 401-412, May 1991, doi:10.1109/34.134040
Usage of this product signifies your acceptance of the Terms of Use.