Matching and motion estimation of three-dimensional point and line sets using eigenstructure without correspondences

Abstract This paper presents moment-based algorithms for matching and motion estimation of three-dimensional (3D) point or line sets without correspondences and application of these algorithms to object tracking over long image sequences. The motion analysis is performed by identifying two sets of coordinate directions based on the relative position of points (or lines) before and after the motion. Since these coordinate vectors are motion invariant, the relationship between them gives parameters of rigid motion. However, we need to determine the set matching before and after the motion estimation algorithms can be applied. We propose several measures suitable for matching of 3D point (and line) sets. We also evaluate these approaches with simulated data and develop criteria for determining the sensitivity to noise. Finally, we apply the proposed algorithm to a time sequence of experimental data (moving vehicle) on which 3D points were determined by stereo matching.

[1]  Hon-Son Don,et al.  A structural approach to finding the point correspondences between two frames , 1988, Proceedings. 1988 IEEE International Conference on Robotics and Automation.

[2]  James Hardy Wilkinson,et al.  Rounding errors in algebraic processes , 1964, IFIP Congress.

[3]  Thomas S. Huang,et al.  Estimation of rigid body motion using straight line correspondences , 1986, Comput. Vis. Graph. Image Process..

[4]  B. Noble.  Applied Linear Algebra , 1969 .

[5]  Olivier D. Faugeras,et al.  A 3-D Recognition and Positioning Algorithm Using Geometrical Matching Between Primitive Surfaces , 1983, IJCAI.

[6]  Alan M. Wood,et al.  Motion analysis , 1986 .

[7]  G. Stewart.  Introduction to matrix computations , 1973 .

[8]  Anup Basu,et al.  A Robust Algorithm for Determining the Translation of a Rigidly Moving Surface without Correspondence, for Robotics Applications , 1987, IJCAI.

[9]  Thomas S. Huang,et al.  Estimating rigid‐body motion from three‐dimensional data without matching point correspondences , 1990, Int. J. Imaging Syst. Technol..

[10]  Thomas S. Huang,et al.  MAXIMAL MATCHING OF TWO THREE-DIMENSIONAL POINT SETS. , 1986 .

[11]  Narendra Ahuja,et al.  Estimating motion/structure from line correspondences: a robust linear algorithm and uniqueness theorems , 1988, Proceedings CVPR '88: The Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  Narendra Ahuja,et al.  3-D Motion Estimation, Understanding, and Prediction from Noisy Image Sequences , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  P. Schönemann.  On artificial intelligence , 1985, Behavioral and Brain Sciences.

[14]  Thomas S. Huang,et al.  FINDING 3-D POINT CORRESPONDENCES IN MOTION ESTIMATION. , 1986 .

[15]  Richard Szeliski.  Estimating Motion From Sparse Range Data Without Correspondence , 1988, [1988 Proceedings] Second International Conference on Computer Vision.

[16]  Minas E. Spetsakis,et al.  Closed Form Solution to the Structure from Motion Problem from Line Correspondences , 1987, AAAI.