Reconstruction from Image Velocities

Part of the Springer Series in Information Sciences book series (SSINF, volume 28)


The velocities of points in an image taken by a moving camera depend on the motion of the camera relative to the environment and on the distances to the object points from which they arise. It is these dependencies that allow the reconstruction of the camera velocity and the shape of the scene from the image velocities. If the full camera calibration is known then reconstruction yields the angular velocity of the camera, the direction of the translational velocity and the shape of the scene up to a single unknown scale factor. Reconstruction from image velocities is a limiting case of reconstruction from image correspondences, as the distances between corresponding points become small. In the limit the underlying equations are simplified, but many of the properties of reconstruction from image correspondences are retained, most notably in the ambiguous case.


Angular Velocity Singular Point Base Point Inverse Distance Optical Centre 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Grzywacz N.M. & Hildreth E.C. 1987 Incremental rigidity scheme for recovering structure from motion: position-based versus velocity-based methods. J. Optical Soc. America, Series A 4, 503–518.Google Scholar
  2. Hardy G.H. 1967 A Course of Pure Mathematics. Cambridge: Cambridge University Press.Google Scholar
  3. Horn B.K.P. 1987 Motion fields are hardly ever ambiguous. International J. Computer Vision 1, 263–278.Google Scholar
  4. Jerian C. & Jain R. 1990 Polynomial methods for structure from motion. IEEE Trans. Pattern Analysis and Machine Intelligence 12, 1150–1166.CrossRefGoogle Scholar
  5. Koenderink J.J. & Van Dorn A.J. 1976 Local structure of movement parallax of the plane. J. Optical Soc. America, 66, 717–723.ADSCrossRefGoogle Scholar
  6. Longuet-Higgins H.C. 1984 The visual ambiguity of a moving plane. Proc. Royal Soc. London, Series B 223, 165–175.ADSMathSciNetCrossRefGoogle Scholar
  7. Longuet-Higgins H.C. & Prazdny K. 1980 The interpretation of a moving retinal image. Proc. Royal Soc. Lond, Series B 208, 385–397.ADSCrossRefGoogle Scholar
  8. Maybank S.J. 1985 The angular velocity associated with the optical flow field due to a rigid moving body. Proc. Royal Soc. London, Series A 401, 317–326.ADSMathSciNetCrossRefGoogle Scholar
  9. Maybank S.J. 1987a Apparent area of a rigid moving body. Image and Vision Computing 15, 111–113.CrossRefGoogle Scholar
  10. Maybank S.J. 1987b A theoretical study of optical flow. PhD thesis, University of London, Birkbeck College.Google Scholar
  11. Maybank S.J. 1990 Rigid velocities compatible with five image velocity vectors. Image and Vision Computing 8, 18–23.CrossRefGoogle Scholar
  12. Murray D.W. & Buxton B.F. 1990 Experiments in the Machine Interpretation of Visual Motion. Cambridge, Massachusetts: The MIT Press.Google Scholar
  13. Semple J.G. & Roth R. 1949 Introduction to Algebraic Geometry. Oxford: Clarendon Press, reprinted 1985.zbMATHGoogle Scholar
  14. Sokolnikoff I.S. & Redheffer R.M. 1966 Mathematics of Physics and Modern Engineering. USA: McGraw-Hill.zbMATHGoogle Scholar
  15. Subbarao M. 1990 Bounds on time-to-collision and rotational component from first-order derivatives of image flow. Computer Vision, Graphics, and Image Processing 50, 329–341.CrossRefGoogle Scholar
  16. Waxman A.M., Kamgar-Parsi B. & Subbarao M. 1987 Closed form solutions to image flow equations for 3D structure and motion. International J. Computer Vision 1, 239–258.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1993

Authors and Affiliations

  1. 1.Hirst Research CentreGEC-Marconi LimitedWembley, MiddlesexUK

Personalised recommendations