A computational framework for determining stereo correspondence from a set of linear spatial filters

  • David G. Jones
  • Jitendra Malik
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 588)


We present a computational framework for stereopsis based on the outputs of linear spatial filters tuned to a range of orientations and scales. This approach goes beyond edge-based and area-based approaches by using a richer image description and incorporating several stereo cues that have previously been neglected in the computer vision literature.

A technique based on using the pseudo-inverse is presented for characterizing the information present in a vector of filter responses. We show how in our framework viewing geometry can be recovered to determine the locations of epipolar lines. An assumption that visible surfaces in the scene are piecewise smooth leads to differential treatment of image regions corresponding to binocularly visible surfaces, surface boundaries, and occluded regions that are only monocularly visible. The constraints imposed by viewing geometry and piecewise smoothness are incorporated into an iterative algorithm that gives good results on random-dot stereograms, artificially generated scenes, and natural grey-level images.


Image Patch Spatial Filter Stereo Match Stereo Pair Filter Response 
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.


  1. Arnold RD, Binford TO (1980) Geometric constraints on stereo vision. Proc SPIE 238:281–292Google Scholar
  2. Ayache N, Faverjon B (1987) Efficient registration of stereo images by matching graph descriptions of edge segments. Int J Computer Vision 1(2):107–131Google Scholar
  3. Baker HH, Binford TO (1981) Depth from edge-and intensity-based stereo. Proc 7th IJCAI 631–636Google Scholar
  4. Barnard ST, Thompson WB (1980) Disparity analysis of images. IEEE Trans PAMI 2(4):333–340Google Scholar
  5. Burt P, Julesz B (1980) A disparity gradient limit for binocular function. Science 208:651–657Google Scholar
  6. DeValois R, DeValois K (1988) Spatial vision. Oxford Univ PressGoogle Scholar
  7. Faugeras O, Maybank S (1990) Motion from point matches: multiplicity of solutions. Int J Computer Vision 4:225–246Google Scholar
  8. Freeman WT, Adelson EH (1991) The design and use of steerable filters. IEEE Trans PAMI 13(9):891–906Google Scholar
  9. Gennery DB (1977) A stereo vision system for autonomous vehicles. Proc 5th IJCAI 576–582Google Scholar
  10. Gillam B, Lawergren B (1983) The induced effect, vertical disparity, and stereoscopic theory. Perception and Psychophysics 36:559–64Google Scholar
  11. Golub GH, Van Loan CF (1983) Matrix computations. The Johns Hopkins Univ Press, Baltimore, MDGoogle Scholar
  12. Grimson WEL (1981) From images to surfaces. M.I.T Press, Cambridge, MassGoogle Scholar
  13. Hannah MJ (1974) Computer matching of areas in images. Stanford AI Memo #239Google Scholar
  14. Hoff W, Ahuja N (1989) Surfaces from stereo: integrating stereo matching, disparity estimation and contour detection. IEEE Trans PAMI 11(2):121–136Google Scholar
  15. Jones, DG (1991) Computational models of binocular vision. PhD Thesis, Stanford UnivGoogle Scholar
  16. Jones DG, Malik J (1991) A computational framework for determining stereo correspondence from a set of linear spatial filters. U.C. Berkeley Technical Report UCB-CSD 91-655Google Scholar
  17. Jones DG, Malik J (1992) Determining three-dimensional shape from orientation and spatial frequency disparities. Proc ECCV, GenovaGoogle Scholar
  18. Kass M (1983) Computing visual correspondence. DARPA IU Workshop 54–60Google Scholar
  19. Kemp M (Ed) (1989) Leonardo on painting. Yale Univ. Press: New Haven 65–66Google Scholar
  20. Koenderink JJ, van Doom AJ (1987) Representation of local geometry in the visual system. Biol Cybern 55:367–375PubMedGoogle Scholar
  21. Koenderink JJ (1988) Operational significance of receptive field assemblies. Biol Cybern 58:163–171PubMedGoogle Scholar
  22. Marr D, Poggio T (1979) A theory for human stereo vision. Proc Royal Society London B 204:301–328Google Scholar
  23. Mayhew JEW (1982) The interpretation of stereo disparity information: the computation of surface orientation and depth. Perception 11:387–403PubMedGoogle Scholar
  24. Mayhew JEW (1983) Stereopsis. in Physiological and Biological Processing of Images. Braddick OJ, Sleigh AC (Eds) Springer-Verlag, Berlin.Google Scholar
  25. Medioni G, Nevatia R (1985) Segment-based stereo matching. CVGIP 31:2–18Google Scholar
  26. Moravec HP (1977) Towards automatic visual obstacle avoidance. Proc 5th IJCAIGoogle Scholar
  27. Nakayama K, Shimojo S (1990) DaVinci Stereopsis: Depth and subjective occluding contours from unpaired image points Vision Research 30(11):1811–1825PubMedGoogle Scholar
  28. Perona P (1991) Deformable kernels for early vision. IEEE Proc CVPR 222–227Google Scholar
  29. Pollard SB, Mayhew JEW, Frisby JP (1985) PMF: a stereo correspondence algorithm using a disparity gradient limit. Perception 14:449–470PubMedGoogle Scholar
  30. Young R (1985) The Gaussian derivative theory of spatial vision: analysis of cortical cell receptive field line-weighting profiles. General Motors Research TR #GMR-4920Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • David G. Jones
    • 1
  • Jitendra Malik
    • 2
  1. 1.Dept. of Electrical EngineeringMcGill UniversityMontréalCanada
  2. 2.Computer Science DivisionUniversity of California, BerkeleyBerkeleyUSA

Personalised recommendations