Title :
Monocular optical flow for real-time vision systems
Author :
Benoit, Stephen M. ; Ferrie, Frank P.
Author_Institution :
Res. Centre for Intelligent Machines, McGill Univ., Montreal, Que., Canada
Abstract :
This paper introduces a monocular optical flow algorithm that has been shown to perform well at nearly real-time frame rates (4 FPS on a 100 MHz SGI Indy workstation), using natural image sequences. The system is completely bottom-up, using pixel region-matching techniques. A coordinated gradient descent method is broken down into two stages: pixel region matching error measures are locally minimized and flow field consistency constraints apply nonlinear adaptive diffusion, causing confident measurements to influence their less confident neighbors. Convergence is usually accomplished with one iteration for an image frame pair. Temporal integration predicts upcoming flow fields. The algorithm is designed for flexibility: large displacements are tracked as easily as sub-pixel displacements, and higher-level information can feed flow field predictions into the measurement process
Keywords :
computer vision; convergence of numerical methods; image matching; image reconstruction; image segmentation; image sequences; iterative methods; real-time systems; bottom-up method; computer vision; convergence; flow field consistency; gradient descent method; image frame pair; image sequences; iterative method; matching error; monocular optical flow; nonlinear adaptive diffusion; pixel region-matching; real-time vision systems; shape recovery; temporal integration; Algorithm design and analysis; Convergence; Coordinate measuring machines; Fluid flow measurement; Image motion analysis; Image sequences; Machine vision; Nonlinear optics; Real time systems; Workstations;
Conference_Titel :
Pattern Recognition, 1996., Proceedings of the 13th International Conference on
Conference_Location :
Vienna
Print_ISBN :
0-8186-7282-X
DOI :
10.1109/ICPR.1996.546147