![](bg6.png)
5 CONCLUSIONS AND FUTURE
WORK
In this paper we have presented a novel Fourier
domain technique to estimate the affine apparent
motion between two views that only needs
corresponding contours. Our technique does not
need explicit point-to-point correspondence. The
normalization of the contours based on the affine arc
length was indispensable when the movement is
assumed affine.
Experiments have shown the applicability of our
technique to a variety of real world problems.
In a future work we would extend our method for
projective homography estimation.
Further experiments would be carried out to
validate our experiments. Parameters such as the
number of points, noise (discretization, sampling,
localization etc.), symmetry in contour, occlusion,
etc. can affect the performance of the proposed
algorithm. An analysis with respect to these
parameters can prove further the performance of the
proposed technique.
REFERENCES
Agarwal, A., Jawahar, C. V. and Narayanan, P. J. A
Survey of Planar Homography Estimation Techniques.
IIIT Technical Report IIIT/TR/2005/12, June 2005.
Ghorbel, F., Mokadem, A., Daoudi, M., Avaro, O. and
Sanson, H. Global planar rigid motion estimation
applied to object-oriented coding, Conf. Proc. ICPR
96, vol. I, 1996, pp. 641-645.
Ghorbel. F. Towards a unitary formulation for invariant
images description: application to images coding,
Annales des Télécommunications, France 1998.
Hartley, R.I. and Zisserman, A. Multiple View Geometry
inComputer Vision, 2nd Edition. Cambridge
University Press, 2004.
Kanatani, K. Statistical Optimization for Geometric
Computation: Theory and Practice. Elsevier Science,
1996.
Kruger, S. and Calway, A. Image Registration Using
Multiresolution Frequency Domain Correlation. In
British Machine Vision Conference (BMVC), pages
316.325, September 1998.
Kumar, M. P., Goyal, S., Kuthirummal, S., Jawahar, C. V.
and Narayanan, P. J. Discrete Contours in Multiple
Views: Approximation and Recognition. Journal of
Image and Vision Computing (IVC),
22(14):1229.1239, December 2004.
Kumar, M. P., Jawahar, C. V. and Narayanan, P. J.
Geometric Structure Computation from Conics. In
Proceedings of Indian Conference on Computer
Vision, Graphics, and Image Processing (ICVGIP),
pages 9.14, 2004.
Kumar, P. J., Jawahar, C. V. "Homography Estimation
from Planar Contours," 3dpvt, pp.877-884, Third
International Symposium on 3D Data Processing,
Visualization, and Transmission (3DPVT'06), 2006.
Kuthirummal, S., Jawahar, C. V. and Narayanan, P. J.
Planar Shape Recognition across Multiple Views. In
Proceedings of the International Conference on
Pattern Recognition (ICPR), pages 482.488, August
2002.
Lisani, J. L. Shape based Automatic Images Comparison,
Ph.D. Thesis, University Paris IX-Dauphine, 2001.
Pauwels, E. J., Recognition of planar shapes under affine
distortion, Int. J. Comput. Vision 14(1995) 49-65.
Sugimoto, A., A Linear Algorithm for Computing the
Homography from Conics in Correspondence. Journal
of Mathematical Imaging and Vision, 13:115.130,
2000.
Zuliani, M., Bhagavathy, S., Manjunath, B. S. and
Kenney, C. S. Affine-Invariant Curve matching. 2004.
APPARENT MOTION ESTIMATION USING PLANAR CONTOURS AND FOURIER DESCRIPTORS
327