Rethinking the Prior Model for Stereo
Sometimes called the smoothing assumption, the prior model of a stereo matching algorithm is the algorithm’s expectation on the surfaces in the world. Any stereo algorithm makes assumptions about the probability to see each surface that can be represented in its representation system. Although the past decade has seen much continued progress in stereo matching algorithms, the prior models used in them have not changed much in three decades: most algorithms still use a smoothing prior that minimizes some function of the difference of depths between neighboring sites, sometimes allowing for discontinuities.
However, one system seems to use a very different prior model from all other systems: the human vision system. In this paper, we first report the observations we made in examining human disparity interpolation using stereo pairs with sparse identifiable features. Then we mathematically analyze the implication of using current prior models and explain why the human system seems to use a model that is not only different but in a sense diametrically opposite from all current models. Finally, we propose two candidate models that reflect the behavior of human vision. Although the two models look very different, we show that they are closely related.
Unable to display preview. Download preview PDF.
- 1.Ayache, N.: Artificial Vision for Mobile Robots. MIT Press, Cambridge (1991)Google Scholar
- 3.Belhumeur, P.N., Mumford, D.: A Bayesian treatment of the stereo correspondence problem using half-occluded regions. In: Proc. CVPR 1992, pp. 506–512 (1992)Google Scholar
- 4.Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. IEEE T. PAMI 23, 1222–1239 (2001)Google Scholar
- 5.Faugeras, O.: Three-Dimensional Computer Vision. MIT Press, Cambridge (1993)Google Scholar
- 8.Grimson, W.E.: From Images to Surfaces. MIT Press, Cambridge (1981)Google Scholar
- 9.Ishikawa, H., Geiger, D.: Occlusions, discontinuities, and epipolar lines in stereo. In: Fifth European Conference on Computer Vision, Freiburg, Germany, pp. 232–248 (1998)Google Scholar
- 10.Jones, J., Malik, J.: Image Vision Comput. 10, 699–708 (1992)Google Scholar
- 11.Kanizsa, G.: Organization in Vision. Praeger, New York (1979)Google Scholar
- 12.Kolmogorov, V., Zabih, R.: Computing Visual Correspondence with Occlusions via Graph Cuts. In: ICCV 2001, Vancouver, Canada, pp. 508–515 (2001)Google Scholar
- 13.Malik, J.: On Binocularly viewed occlusion Junctions. In: Fourth European Conference on Computer Vision, vol. 1, pp. 167–174 (1996)Google Scholar
- 21.Roy, S., Cox, I.: A maximum-flow formulation of the N-camera stereo correspondence problem. In: International Conference on Computer Vision, Bombai, India, pp. 492–499 (1998)Google Scholar
- 23.Szeliski, R.: A Bayesian modelling of uncertainty in low-level vision. Kluwer Academic Press, Boston (1989)Google Scholar