Progress in Trinocular Stereo

  • Matti Pietikäinen
  • Olli Silvén
Conference paper
Part of the NATO ASI Series book series (volume 42)

Abstract

Stereovision is an important technique for obtaining depth information from a scene by triangulation of corresponding points in images taken from different viewpoints. A major problem is to find the corresponding points in the two images. There is no complete solution to this problem, but constraints are needed to reduce the search for a unique solution among possible matches. But even with constraints, the process of eliminating false matches tends to be complex and slow, and many errors may remain. Many of the accepted matches may be spurious, not corresponding to any physical structure in the scene. Spurious matches tend in particular to occur at occluding boundaries, where physical points may be visible in only one of the images. Besides this, in the ordinary stereo, it is impossible to precisely match edges which are parallel or nearly parallel to the epipolar lines.

Keywords

Edge Point Stereo Match Epipolar Line False Match Candidate Match 
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.
This is a preview of subscription content, log in to check access.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. /1/.
    Aloimonos J., Bandyopadhyay A., Chou P., On the Foundations of Trinocular Machine Vision, University of Rochester, Department of Computer Science, Rochester, New York, 1985.Google Scholar
  2. /2/.
    Ayache N., Lustman F., Fast and Reliable Passive Stereovision Using Three Cameras, INRIA, Le Chesnay, France, 1986.Google Scholar
  3. /3/.
    Faugeras O.D., Toscani G., The Calibration Problem for Stereo, Proc. IEEE Conference on Computer Vision and Pattern Recognition, Miami Beach, Florida, 1986, pp. 15–20.Google Scholar
  4. /4/.
    Gerhard A., Platzer H., Steurer J., Lenz R., Depth Extraction by Stereo Triples and a Fast Correspondence Estimation Algorithm, Proc. Eighth International Conference on Pattern Recognition, Paris, France, 1986, pp. 512–515.Google Scholar
  5. /5/.
    Gurewitz E., Dinstein I., Sarusi B., More on the Benefit of a Third Eye for Machine Stereo Perception, Proc. Eighth International Conference on Pattern Recognition, Paris, France, 1986, pp. 966–968.Google Scholar
  6. /6/.
    Ito M., Ishii A., Three-View Stereo Analysis, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. PAMI-8, No. 4, 1986, pp. 524–532.CrossRefGoogle Scholar
  7. /7/.
    Ito M., Ishii A., Range and Shape Measurement Using Three-View Stereo-Analysis, Proc. IEEE Conference on Computer Vision and Pattern Recognition, Miami Beach, Florida, 1986, pp. 9–14.Google Scholar
  8. /8/.
    Kamgar-Parsi B., Eastman R., Calibration of a Stereo System with Small Relative Angles, University of Maryland, Center for Automation Research, CAR-TR-240, 1986.Google Scholar
  9. /9/.
    Milenkovic V.J., Kanade T., Trinocular Vision Using Photometric and Edge Orientation Constraints, Proc. DARPA Image Understanding Workshop, Miami Beach, Florida, 1985, pp. 163–175.Google Scholar
  10. /10/.
    Moravec H.P., Obstacle Avoidance and Navigation in the Real World by a Seeing Robot Rover, Stanford University, Artificial Intelligence Laboratory, AIM-340, 1980.Google Scholar
  11. /11/.
    Ohta Y., Watanabe M., Ikeda K., Improving Depth Map by Right Angled Trinocular Stereo, Proc. Eighth International Conference on Pattern Recognition, Paris, France, 1986, pp. 519–521.Google Scholar
  12. /12/.
    Pietikäinen M., Harwood D., Multiple-Camera Contour Stereo, University of Maryland, Center for Automation Research, CAR-TR-151, 1985.Google Scholar
  13. /13/.
    Pietikäinen M., Harwood D., Depth from Three Camera Stereo, Proc. IEEE Conference on Computer Vision and Pattern Recognition, Miami Beach, Florida, 1986, pp. 2–8.Google Scholar
  14. /l4/.
    Tsai R.Y., An Efficient and Accurate Camera Calibration Technique for 3D Machine Vision, Proc. IEEE Conference on Computer Vision and Pattern Recognition, Miami Beach, Florida, 1986, pp. 364–374.Google Scholar
  15. /15/.
    Tsai R.Y., Multiframe Image Point Matching and 3-D Surface Reconstruction, IEEE Trans. Pattern Analysis and Machine Intelligence, Vol. PAMI-5, No. 2, 1983, PP. 159–174.CrossRefMATHGoogle Scholar
  16. /16/.
    Yachida M., 3-D Data Acquisition by Multiple Views, Proc. Third International Symposium on Robotics Research (Eds. O.D. Faugeras, G. Giralt), Gouvieux, France, 1985.Google Scholar
  17. /17/.
    Yachida M., Kitamura Y., Kimachi M., Trinocular Vision: New Approach for Correspondence Problem, Proc. Eighth International Conference on Pattern Recognition, Paris, France, 1986, pp. 1041–1044.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • Matti Pietikäinen
    • 1
  • Olli Silvén
    • 1
  1. 1.Department of Electrical EngineeringUniversity of OuluOuluFinland

Personalised recommendations