Trinocular stereo analysis of optical flow

A method for analyzing optical flows in orthographic projections of 3D rigid motions is described. A trinocular stereo constraint for optical flow derived from geometrical relations between three views is proposed. Instead of taking the correspondence between tokens in stereo images, the method is to reconstruct the motion and structure of a rigid motion by solving a nonlinear optimization problem for optical flow from three views with the trinocular stereo constraint. In the case of multiple motions, optical flow from each view is segmented into component optical flows by utilizing the properties of parallel optical flow vectors. The next step is to estimate globally the degree of correspondence between three component optical flows selected from three different views by using the trinocular stereo constraint, and thus to determine plausible correspondence. The optimization process for a single motion reconstructs the 3D information using the established correspondence.<<ETX>>

[1]  Gilad Adiv,et al.  Determining Three-Dimensional Motion and Structure from Optical Flow Generated by Several Moving Objects , 1985, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Johan D'Haeyer.  Determining motion of image curves from local pattern changes , 1986, Comput. Vis. Graph. Image Process..

[3]  Amar Mitiche,et al.  Three-dimensional space from optical flow correspondence , 1988, Comput. Vis. Graph. Image Process..

[4]  Jake K. Aggarwal,et al.  Analyzing orthographic projection of multiple 3D velocity vector fields in optical flow , 1988, Comput. Vis. Graph. Image Process..

[5]  Allen M. Waxman,et al.  Surface Structure and Three-Dimensional Motion from Image Flow Kinematics , 1985 .

[6]  E. Hildreth.  The computation of the velocity field , 1984, Proceedings of the Royal Society of London. Series B. Biological Sciences.

[7]  Allen M. Waxman,et al.  Dynamic Stereo: Passive Ranging to Moving Objects from Relative Image Flows , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  W Richards,et al.  Structure from stereo and motion. , 1985, Journal of the Optical Society of America. A, Optics and image science.

[9]  Berthold K. P. Horn,et al.  Determining Optical Flow , 1981, Other Conferences.

[10]  K. Kanatani.  Structure and motion from optical flow under orthographic projection , 1986 .

[11]  Allen M. Waxman,et al.  Contour Evolution, Neighborhood Deformation, and Global Image Flow: Planar Surfaces in Motion , 1985 .

[12]  N. Otsu.  A threshold selection method from gray level histograms , 1979 .

[13]  Allen M. Waxman,et al.  Binocular Image Flows: Steps Toward Stereo-Motion Fusion , 1986, IEEE Transactions on Pattern Analysis and Machine Intelligence.