Abstract
Binocular stereo is the process of obtaining depth information from a pair of cameras. In the past, stereo algorithms have had problems at occlusions and have tended to fail there (though sometimes post-processing has been added to mitigate the worst effects). We show that, on the contrary, occlusions can help stereo computation by providing cues for depth discontinuities.
We describe a theory for stereo based on the Bayesian approach, using adaptive windows and a prior weak smoothness constraint, which incorporates occlusion. Our model assumes that a disparity discontinuity, along the epipolar line, in one eyealways corresponds to an occluded region in the other eye thus, leading to anocclusion constraint. This constraint restricts the space of possible disparity values, thereby simplifying the computations. An estimation of the disparity at occluded features is also discussed in light of psychophysical experiments. Using dynamic programming we can find the optimal solution to our system and the experimental results are good and support the assumptions made by the model.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Ayache, N.Artificial Vision for Mobile Robots, The MIT Press, Cambridge, Mass., 02142, 1991.
Baker, H.H. and Binford, T.O. Depth from edge and intensity based stereo, InProceedings IJCAI, pp. 631–636, Vancouver, 1981.
Belhumeur, P.N. and Mumford, D. A bayesian treatment of the stereo correspondence problem using half-occluded regions, InProceedings IEEE Conf. on Computer Vision and Pattern Recognition, 1992.
Bellman, R.E.Dynamic Programming, Princeton University Press, 1957.
Blake, A. and Zisserman, A.Visual Reconstruction, MIT Press, Cambridge, Mass., 1987.
Burt, P. and Julesz, B. A disparity gradient limit for binocular fusion,Science, 208:615–617, 1980.
Cernushi-Frias, B., Cooper, D.B., Hung, Y.-P. and Belhumeur, P. Towards a model-based bayesian theory for estimating and recognizing parameterized 3d objects using two or more images taken from different positions,IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-11:1028–1052, 1989.
Champolle, A., Geiger, D. and Mallat, S. Un algorithme multi-échelle de raise encorrespondance stéréo basé sur les champs markoviens, In13th GRETSI Conference on Signal and Image Processing, Juan-les-Pins, France, Sept. 1991.
Drumheller, M. and Poggio, T. On parallel stereo, InProceedings of IEEE Conference on Robotics and Automation, pp. 1439–1448, Washington, DC, 1986, IEEE.
Fuh, C.S. and Maragos, P. Motion displacement estimation using an affine model for image matching,Optical Engineering, 30:881–887, July 1991.
Geiger, D. and Girosi, E. Parallel and deterministic algorithms for turfs: surface reconstruction,IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-13(5):401–412, May 1991.
Geman, S. and Geman, D. Stochastic relaxation, gibbs distributions, and the bayesian restoration of images,IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-6:721–741, 1984.
Gillam, B. and Borsting, E. The role of monocular regions in stereoscopic displays,Perception, 17:603–608, 1988.
Grimson, W.E.L.From Images to Surfaces, MIT Press, Cambridge, Mass., 1981.
Gruen, A.W. Adaptive least squares correlation: a powerful image matching technique,S. Afr. J. of Photogrammetry, Remote Sensing and Cartography, 3(14): 175–187, 1985.
Julesz, B.Foundations of Cyclopean Perception, The University of Chicago Press, Chicago, 1971.
Kanade, T. and Okutomi, M. A stereo matching algorithm with an adaptive window: theory and experiments, InProc. Image Understanding Workshop DARPA, PA, September 1990.
Krol, J.D. and Vander Grind, W.A. The double-nail illusion: experiments on binocular vision with nails, needles and pins,Perception, 11:615–619, 1982.
Little, J. and Gillett, W. Direct evidence for occlusions in stereo and motion, In1st. ECCV, pp. 336–340, Antibes, France, April 1990, Springer-Verlag.
Marr, D. and Poggio, T. Cooperative computation of stereo disparity,Science, 194:283–287, 1976.
Marr, D. and Poggio, T. A computational theory of human stereo vision,Proceedings of the Royal Society of London B, 204:301–328, 1979.
Mumford, D. and Shah, J. Boundary detection by minimizing functionals, i. InProc. IEEE Conf. on Computer Vision & Pattern Recognition, San Francisco, CA, 1985.
Nakayama, K. and Shimojo, S. Da vinci stereopsis: depth and subjective occluding contours from unpaired image points,Vision Research, 30:1811–1825, 1990.
Ohta, Y. and Kanade, T. Stereo by intra- and inter-scanline search using dynamic programming,IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-7(2):139–154, 1985.
Panum, E.L. Physiologische untersuchungen ueber dos sehen mit zwei augen,Kiel, Homann, 1858.
Pollard, S.B., Mayhew, J.E.W. and Frisby, J.E. Disparity gradients and stereo correspondences,Perception, 1987.
Sperling, G. Binocular vision: A physical and a neural theory,American Journal of Psychology, 83:461–534, 1967.
Weinshall, D. Perception of multiple transparent planes in stereo vision,Nature, 341:737–739, 1989.
Yang, Y., Yuille, A.L. and Liu, J. Geometric distortion and stereo matching, Harvard Rob. Lab. Tech. Report. In preparation, Harvard, 1992.
Yuille, A., Geiger, D. and Bulthoff, H. Stereo, mean field theory and psychophysics, In1st. ECCV, pp. 73–82, Antibes, France, April 1990, Springer-Verlag.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Geiger, D., Ladendorf, B. & Yuille, A. Occlusions and binocular stereo. Int J Comput Vision 14, 211–226 (1995). https://doi.org/10.1007/BF01679683
Issue Date:
DOI: https://doi.org/10.1007/BF01679683