Abstract
Stereo matching is an ill-posed problem for at least two principal reasons: (1) because of the random nature of match similarity measure and (2) because of structural ambiguity due to repetitive patterns. Both ambiguities require the problem to be posed in the regularization framework. Continuity is a natural choice for a prior model. But this model may fail in low signal-to-noise ratio regions. The resulting artefacts may then completely spoil the subsequent visual task.
A question arises whether one could (1) find the unambiguous component of matching and, simultaneously, (2) identify the ambiguous component of the solution and then, optionally, (3) regularize the task for the ambiguous component only. Some authors have already taken this view. In this paper we define a new stability property which is a condition a set of matches must satisfy to be considered unambiguous at a given confidence level. It turns out that for a given matching problem this set is (1) unique and (2) it is already a matching. We give a fast algorithm that is able to find the largest stable matching. The algorithm is then used to show on real scenes that the unambiguous component is quite dense (10–80%) and error-free (total error rate of 0.3–1.4%), both depending on the confidence level chosen.
Chapter PDF
Similar content being viewed by others
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Julesz, B.: Towards the automation of binocular depth perception (AUTOMAP-1). In: IFIPS Congress, Munich (1962)
Kutulakos, K.N., Seitz, S.M.: A theory of shape by shape carving. IJCV 38 (2000) 199–218
Baker, S., Sim, T., Kanade, T.: A characterization of inherent stereo ambiguities. In: Proc ICCV. (2001) 428–435
Schlesinger, M.I.: Ambiguity in stereopsis. Personal communication (1998)
Šára, R.: Failure analysis of stable matchings. Research report (In preparation)
Stewart, C.V., Dyer, C.R.: The trinocular general support algorithm: A three-camera stereo algorithm for overcoming binocular matching errors. In: Proc ICCV. (1988) 134–138
Satoh, K., Ohta, Y.: Occlusion detectable stereo using a camera matrix. In: Proc ACCV. (1995) 331–335
Marr, D.: A note on the computation of binocular disparity in a symbolic, low-level visual processor. A.I. Memo 327, AI Lab, MIT (1974)
Pollard, S.B., Mayhew, J.E.W., Frisby, J.P.: PMF: A stereo correspondence algorithm using a disparity gradient limit. Perception 14 (1985) 449–470
Zitnick, C.L., Kanade, T.: A cooperative algorithm for stereo matching and occlusion detection. IEEE Trans PAMI 22 (2000) 675–684
Belhumeur, P.N.: A Bayesian approach to binocular stereopsis. IJCV 19 (1996) 237–260
Bobick, A.F., Intille, S.S.: Large occlusion stereo. IJCV 33 (1999) 181–200
Robert, L., Deriche, R.: Dense depth map reconstruction: A minimization and regularization approach which preserves discontinuities. In: Proc ICIP. (1992) 123–127
Barnard, S.T.: Stochastic stereo matching over scale. IJCV 3 (1989) 17–32
Scharstein, D., Szeliski, R.: Stereo matching with nonlinear diffusion. IJCV 28 (1998) 155–174
Boykov, Y., Veksler, O., Zabih, R.: Disparity component matching for visual correspondence. In: Proc Conf CVPR. (1997) 470–475
Ishikawa, H., Geiger, D.: Occlusions, discontinuities, and epipolar lines in stereo. In: ECCV. (1998) 232–248
Roy, S., Cox, I.J.: A maximum-flow formulation of the n-camera stereo correspondence problem. In: Proc ICCV. (1998) 492–499
Kolmogorov, V., Zabih, R.: Computing visual correspondence with occlusions using graph cuts. In: Proc ICCV. (2001) 508–515
March, R.: Computation of stereo disparity using regularization. Pattern Recognition Letters 8 (1988) 181–187
March, R.: A regularization model for stereo vision with controlled continuity. Pattern Recognition Letters 10 (1989) 259–263
Jung, D.Y., Oh, J.H., Lee, S.C., Lee, C.H., Nam, K.G.: Stereo matching by discontinuity-preserving regularization. J of Elect Eng and Inf Sci 4 (1999) 452–8
Manduchi, R.; Tomasi, C.: Distinctiveness maps for image matching. In: Proc ICIAP. (1999) 26–31
Šára, R.: Sigma-delta stable matching for computational stereopsis. Research Report CTU-CMP-2001-25, Center for Machine Perception, Czech Technical University (2001) [ ftp://cmp.felk.cvut.cz/pub/cmp/articles/sara/Sara-TR-2001-25.pdf ].
Krol, J.D., van der Grind, W.A.: Rehabilitation of a classical notion of Panum’s fusional area. Perception 11 (1982) 615–619
Yuille, A.L., Poggio, T.: A generalized ordering constraint for stereo correspondence. A.I. Memo 777, AI Lab, MIT (1984)
Šára, R.: A fast algorithm for confidently stable matching. Research Report CTU-CMP-2002-03, Center for Machine Perception, Czech Technical University (2002) [ftp://cmp.felk.cvut.cz/pub/cmp/articles/sara/Sara-TR-2002-03.pdf].
Moravec, H.P.: Towards automatic visual obstacle avoidance. In: Proc IJCAI. (1977) 584
Mandelbaum, R., Kamberova, G., Mintz, M.: Stereo depth estimation: a confidence interval approach. In: Proc ICCV. (1998) 503–509
Cox, I.J., Hingorani, S., Maggs, B.M., Rao, S.B.: Stereo without disparity gradient smoothing: a Bayesian sensor fusion solution. In: Proc BMVC. (1992) 337–346
Scheibe, K., Korsitzky, H., Reulke, R., Scheele, M., Solbrig, M.: EYESCAN-a high resolution digital panoramic camera. In: Int Wkshp Robot Vision, Auckland (2001) 77–83
Scharstein, D., Szeliski, R.: A taxonomy and evaluation of dense two-frame stereo correspondence algorithms. Technical Report MSR-TR-2001-81, Microsoft Research, Redmont, WA (2001) To appear in IJCV.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Šára, R. (2002). Finding the Largest Unambiguous Component of Stereo Matching. In: Heyden, A., Sparr, G., Nielsen, M., Johansen, P. (eds) Computer Vision — ECCV 2002. ECCV 2002. Lecture Notes in Computer Science, vol 2352. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-47977-5_59
Download citation
DOI: https://doi.org/10.1007/3-540-47977-5_59
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-43746-8
Online ISBN: 978-3-540-47977-2
eBook Packages: Springer Book Archive