Shape distances for contour tracking and motion estimation

The aim of this paper is to present two kinds of shape distances in a dynamic image context. Both are defined for closed curves. The first distance is obtained by invariant descriptors, under rigid motion, which verify a completeness and stability properties. The second one distance obtained by a Hausdorff distance that allows to estimate the motion parameters. These distances are tested on real images.

[1]  F. Ghorbel,et al.  Global planar rigid motion estimation applied to object-oriented coding , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[2]  Andrew Blake,et al.  Affine-invariant contour tracking with automatic control of spatiotemporal scale , 1993, 1993 (4th) International Conference on Computer Vision.

[3]  Faouzi Ghorbel.  Stability of invariant Fourier descriptors and its inference in the shape classification , 1992, Proceedings., 11th IAPR International Conference on Pattern Recognition. Vol. III. Conference C: Image, Speech and Signal Analysis,.

[4]  Daniel P. Huttenlocher,et al.  Computing the minimum Hausdorff distance for point sets under translation , 1990, SCG '90.

[5]  Thomas S. Huang,et al.  Image sequence analysis , 1981 .

[6]  Faouzi Ghorbel.  Towards a unitary formulation for invariant image description: application to image coding , 1998, Ann. des Télécommunications.

[7]  Faouzi Ghorbel,et al.  A shape distance by complete and stable invariant descriptors for contour tracking , 1996, Proceedings of 13th International Conference on Pattern Recognition.