Abstract
This paper deals with the problem of depth recovery or surface reconstruction from sparse and noisy range data. The image is modelled as a Markov Random Field and a new potential function is developed to effectively detect discontinuities in highly sparse and noisy images. No use of any line process is made. Interpolation over missing data sites is first done using local characteristics and simulated annealing is then used to compute the maximum a posteriori (map) estimate. Results of software simulations carried out on actual range images along with details of the simulations are presented.
Similar content being viewed by others
References
Aarts E, Korst J 1989Simulated annealing and Boltzmann machines (Chichester: Wiley)
Bertero M, Poggio T, Torre V 1988 Ill-posed problems in early vision.Proc. IEEE 76: 869–889
Besag J E 1974 Spatial interaction and statistical analysis of lattice systems.J. R. Stat. Soc. B-36: 192–236
Besag J E 1977 On the statistical analysis of dirty pictures.J. R. Stat. Soc. B-48: 259–302
Bolle R M, Vemuri B C 1991 On three-dimensional surface reconstruction methods.IEEE Trans. Pattern Anal. Machine Intell. PAMI-13: 1–13
Boult T E, Kender J R 1986 Visual surface reconstruction using sparse depth data.Proc. IEEE Conf. Comput. Vision Pattern Recog. (New York: ieee)
Choi D J, Kender J R 1988 Solving the depth interpolation problem on a parallel architecture with multigrid approach.Proc. IEEE Conf. Comput. Vision Pattern Recog. (New York: ieee)
Frankot R T, Chellappa R 1988 A method for enforcing integrability in shape from shading algorithms.IEEE Trans. Pattern Anal. Machine Intell. PAMI-10: 439–451
Geiger D, Girosi F 1991 Parallel and deterministic algorithms from mrfs: Surface reconstruction.IEEE Trans. Pattern Anal. Machine Intell. PAMI-13: 401–412
Geman S, Geman D 1984 Stochastic relaxation, Gibbs distributions and the Bayesian distribution of images.IEEE Trans. Pattern Anal. Machine Intell. PAMI-6: 721–741
Geman S, Graffigne C 1987 Markov random fields image models and their applications to computer vision.Proc. Int. Congr. Math. (ed.) A M Gleason (Providence, ri: Am. Math. Soc.)
Horn B K P, Brooks M J (eds) 1989Shape from shading (Cambridge, ma: mit Press)
Kapoor S, Mundkur P Y, Desai U B 1992 Depth and image recovery using a mrf model. Technical Report, Dept. Elec. Eng., Indian Institute of Technology, Bombay
Kirkpatrick S, Gelatt G D, Vecchi M P 1983 Optimization by simulated annealing.Science 220: 671–680
Lakshmanan S, Derin H 1989 Simultaneous parameter estimation and segmentation of Gibbs random fields using simulated annealing.IEEE Trans. Pattern Anal. Machine Intell. PAMI-11: 799–813
Marroquin J, Mitter S, Poggio T 1987 Probabilistic solution of ill-posed problems in computational vision.J. Am. Stat. Assoc. 82: 76–89
Mundkur P Y, Kapoor S, Desai U B 1992 Boltzmann machines for depth recovery using a mrf model.Proc. Int. Joint Conf. Neural Networks, Baltimore
Prasad K V V, Mammone R J, Yogeshwar J 1990 3-D restoration using constrained optimization techniques.Opt. Eng. special issue onImage Restoration and Reconstruction
Shao M, Simchony T, Chellappa R 1987 New algorithms for reconstruction of a 3-D depth map from one or more images.Proc. IEEE Conf. Comput. Vision Pattern Recog. (New York: ieee)
Terzopoulos D 1988 The computation of visible surface representation.IEEE Trans. Pattern Anal. Machine Intell. PAMI-10: 417–438
Younes L 1988 Estimation and annealing for Gibbsian fields.Ann. Inst. Henri Poincaré 24: 269–294
Zerubia J, Chellappa R 1990 Mean field annealing for edge detection and image restoration.Proc. European Signal Proc. Conf. Barcelona, Spain
Author information
Authors and Affiliations
Additional information
This work was supported by the mhrd grants for Laboratory for Artificial Neural Networks and for Digital Communication and Signal Processing.
Rights and permissions
About this article
Cite this article
Kapoor, S., Mundkur, P.Y. & Desai, U.B. Depth recovery using Markov Random Fields. Sadhana 18, 17–29 (1993). https://doi.org/10.1007/BF02811384
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02811384