Identifying multiple motions from optical flow

  • Alessandra Rognone
  • Marco Campani
  • Alessandro Verri
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 588)


This paper describes a method which uses optical flow, that is, the apparent motion of the image brightness pattern in time-varying images, in order to detect and identify multiple motions. Homogeneous regions are found by analysing local linear approximations of optical flow over patches of the image plane, which determine a list of the possibly viewed motions, and, finally, by applying a technique of stochastic relaxation. The presented experiments on real images show that the method is usually able to identify regions which correspond to the different moving objects, is also rather insensitive to noise, and can tolerate large errors in the estimation of optical flow.


Vector Field Image Plane Optical Flow Apparent Motion Real Image 
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. 1.
    Adiv, G. Determining three-dimensional motion and structure from optical flow generated by several moving objects. IEEE Trans. Pattern Anal. Machine Intell. 7 (1985), 384–401.Google Scholar
  2. 2.
    Francois, E. and P. Bouthemy. Derivation of qualitative information in motion analysis. Image and Vision Computing 8 (1990), 279–288.Google Scholar
  3. 3.
    Gibson, J.J. The perception of the Visual World. (Boston, Houghton Mifflin, 1950).Google Scholar
  4. 4.
    Koenderink, J.J. and Van Dorn, A.J. How an ambulant observer can construct a model of the environment from the geometrical structure of the visual inflow. In Kibernetic 1977, G. Hauske and E. Butendant (Eds.), (Oldenbourg, Munchen, 1977).Google Scholar
  5. 5.
    Verri, A., Girosi, F., and Torre, V. Mathematical properties of the two-dimensional motion field: from Singular Points to Motion Parameters. J. Optical Soc. Amer. A 6 (1989), 698–712.Google Scholar
  6. 6.
    Subbarao, M. Bounds on time-to-collision and rotational component from first order derivatives of image flow. CVGIP 50 (1990), 329–341.Google Scholar
  7. 7.
    Campani, M., and Verri, A. Motion analysis from first order properties of optical flow. CVGIP: Image Understanding in press (1992).Google Scholar
  8. 8.
    Bouthemy, P. and Santillana Rivero, J. A hierarchical likelihood approach for region segmentation according to motion-based criteria. In Proc. 1st Intern. Conf. Comput. Vision London (UK) (1987), 463–467.Google Scholar
  9. 9.
    Adiv, G. Inherent ambiguities in recovering 3D motion and structure from a noisy flow field. Pattern Anal. Machine Intell. 11 (1989), 477–489.Google Scholar
  10. 10.
    Campani, M. and A. Verri. Computing optical flow from an overconstrained system of linear algebraic equations. Proc. 3rd Intern. Conf. Comput. Vision Osaka (Japan) (1990), 22–26.Google Scholar
  11. 11.
    Geman, D., Geman, S., Graffigne, C., and P. Dong. Boundary detection by constrained optimization. IEEE Trans. Pattern Anal. Machine Intell. 12 (1990), 609–628.Google Scholar
  12. 12.
    Thompson, W.B. and Ting Chuen Pong. Detecting moving objects. IJCV 4 (1990), 39–58.Google Scholar
  13. 13.
    Verri, A., Girosi, F., and Torre, V. Differential techniques for optical flow. J. Optical Soc. Amer. A 7 (1990), 912–922.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Alessandra Rognone
    • 1
  • Marco Campani
    • 1
  • Alessandro Verri
    • 1
  1. 1.Dipartimento di Fisica dell'Università di GenovaGenovaItaly

Personalised recommendations