Distributed Computation of Optical Flow

This paper describes a new parallel algorithm to compute the optical flow of a video sequence. A previous sequential algorithm has been distributed over a cluster. It has been implemented in a cluster with 8 nodes connected by means of a Gigabit Ethernet. On this architecture, the algorithm, that computes the optical flow of every image on the sequence, is able of processing 10 images of 720x576 pixels per second.

[1]  M.V. Correia,et al.  Optical flow techniques applied to the calibration of visual perception experiments , 1996, Proceedings of 13th International Conference on Pattern Recognition.

[2]  Eero P. Simoncelli.  Distributed representation and analysis of visual motion , 1993 .

[3]  B. D. Lucas.  Generalized image matching by the method of differences , 1985 .

[4]  Abbas El Gamal,et al.  Optical flow estimation using high frame rate sequences , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[5]  Aurélio J. C. Campilho,et al.  Real-time implementation of an optical flow algorithm , 2002, Object recognition supported by user interaction for service robots.

[6]  Takeo Kanade,et al.  An Iterative Image Registration Technique with an Application to Stereo Vision , 1981, IJCAI.

[7]  Keith Langley,et al.  Recursive Filters for Optical Flow , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  David J. Fleet,et al.  Performance of optical flow techniques , 1994, International Journal of Computer Vision.

[9]  John Zelek,et al.  Bayesian Real-time Optical Flow , 2002 .

[10]  B. Crespi,et al.  Real-time parallel computation of disparity and optical flow using phase difference , 1996 .

[11]  Edward H. Adelson,et al.  Probability distributions of optical flow , 1991, Proceedings. 1991 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[12]  Timo Kohlberger,et al.  Domain Decomposition for Parallel Variational Optical Flow Computation , 2003, DAGM-Symposium.