Subpixel Flow Detection by the Hough Transform
In this paper, we show that randomized sampling and voting processes allow to treat linear flow field detection as a model-fitting problem. If we use an appropriate number of images from a sequence of images, it is possible to detect subpixel motion in this sequence. We use the accumulator space for the unification of these flow vectors which are computed from different time intervals.
Unable to display preview. Download preview PDF.
- 3.E. Oja, L. Xu, P. Kultanen: Curve detection by an extended self-organization map and related RHT method. Proceed. Internat. on Neural Network Conf., 1 (1990) 27–30. 140Google Scholar
- 4.H. Kälviäinen, E. Oja, L. Xu: Randomized Hough transform applied to translation and rotation motion analysis. 11th IAPR Proceed. of Internat. Conf. on Pattern Recognition (1992) 672–675. 140Google Scholar
- 6.K. Kanatani: Statistical optimization and geometric inference in computer vision, Philosopical Transactions of the Royal Society of London, Series A, 356 (1997) 1308–1320.Google Scholar
- 8.J. L. Barron, D. J. Fleet, S. S. Beauchemin: Performance of optical flow techniques. Report No. 299, Department of Computer Science, The University of Western Ontario, London (1992). 142, 145Google Scholar