Abstract
The paper presents a novel approach to the Robust Analysis of Complex Motion. It employs a low-level robust motion estimator, conceptually based on the Hough Transform, and uses Multiresolution Markov Random Fields for the global interpretation of the local, low-level estimates. Motion segmentation is performed in the front-end estimator, in parallel with the motion parameter estimation process. This significantly improves the accuracy of estimates, particularly in the vicinity of motion boundaries, facilitates the detection of such boundaries, and allows the use of larger regions, thus improving robustness. The measurements extracted from the sequence in the front-end estimator include displacement, the spatial derivatives of the displacement, confidence measures, and the location of motion boundaries. The measurements are then combined within the MRF framework, employing the supercoupling approach for fast convergence. The excellent performance, in terms of estimate accuracy, boundary detection and robustness is demonstrated on synthetic and real-word sequences.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
J.L. Barron, D.J. Fleet, S.S. Beauchemin, and T.A Burkitt. Performance of optical flow techniques. In Proceedings CVPR'92 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Champaign, IL., 1992), pages 236–242, Washington, DC., 1992. Computer Society Press.
M. Bober and J. Kittler. General motion estimation and segmentation. In Proceedings Conf Pattern Recognition and Image Analysis, Valencia, Spain, September 1992, pages 10–18, 1992.
M. Bober and J. Kittler. Estimation of general multimodal motion: an approach based on robust statistics and Hough transform. Image and Vision Computing, 12(12):661–668, 1994.
M. Bober and J. Kittler. Robust motion analysis. In Proceedings, CVPR '94 (IEEE Computer Society Conference on Computer Vision and Pattern Recognition), Seattle, June 20–24, 1994, pages 947–952. IEEE Computer Society Press, 1994.
F. Heitz and P. Bouthemy. Multimodal estimation of discontinuous optical flow using markov random fields. IEEE Transactions on Pattern Analysis and Machine Intelligence, 15:1217–1232, 1993.
Proceedings, International Conference on Pattern Recognition, Jerusalem, October 1994. IEEE Computer Society Press, 1994.
G. K. Nicholls and M. Petrou. On multiresolution image restoration. In ICPR1994 [6], pages 63–67.
M. Petrou, M. Bober, and J. Kittler. Multiresolution motion segmentation. In ICPR1994 [6], pages 379–383.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Bober, M., Kittler, J. (1996). Combining the hough transform and multiresolution MRF's for the robust motion estimation. In: Li, S.Z., Mital, D.P., Teoh, E.K., Wang, H. (eds) Recent Developments in Computer Vision. ACCV 1995. Lecture Notes in Computer Science, vol 1035. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-60793-5_65
Download citation
DOI: https://doi.org/10.1007/3-540-60793-5_65
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-60793-9
Online ISBN: 978-3-540-49448-5
eBook Packages: Springer Book Archive