On the Computation of Optical Flow using the 3-D Gabor Transform
The motion of brightness patterns in an image sequence (optical flow) is most intuitively considered as a spatiotemporal phenomenon. It has been shown, however, that motion has a characteristic signature in the spatiotemporal-frequency (Fourier) domain. This fact can be exploited for the computation of optical flow. However, for cases which involve a number of regions in a sequence with different motions, as in scenes with one or more objects moving against a stationary or moving background, the global nature of the Fourier transform makes it unsuitable for this task. The signatures of the different motions cannot be resolved in the Fourier domain, nor associated with their respective regions in the image sequence. Local frequency representations provide a means to address this problem. In this paper, we consider the application of a 3-D version of the widely used Gabor transform to the computation of optical flow.
Key Wordsmotion estimation optical flow Gabor transform local frequency representations
Unable to display preview. Download preview PDF.
- 2.A. B. Watson, and A. J. Ahumada, “A Look at Motion in the Frequency Domain,” SIGGRAPH/SIGART Interdisciplinary Workshop MOTION: Representation and Perception, Toronto, Canada, April 4–6 1983, pp. 1–10.Google Scholar
- 6.A. Kojima, N. Sakurai, and J. Kishigami, “Motion Detection Using 3D-FFT Spectrum,” Proc. of IEEE ICASSP ‘83, Minneapolis, Minnesota, April 27–30, pp. 213–216.Google Scholar
- 8.H. Jacobson, and H. Wechsler, “Derivation of Optical Flow Using a Spatiotemporal-Frequency Approach,” Computer Vision, Graphics, and Image Processing, vol. 38, 1987, pp. 29–65.Google Scholar
- 9.D. Gabor, “Theory of Communication,” Proc. of the Inst. of Elec. Eng., vol. 93, no. 26, 1946, pp. 429–457.Google Scholar
- 10.T. Ebrahimi, T. R. Reed, and M. Kunt, “Video Coding Using a Pyramidal Gabor Expansion,” Proc. of VCIP ‘80, Lausanne, Switzerland, October 2–4, pp. 489–502.Google Scholar