Title :
On the accuracy of optical flow computation using global optimization
Author :
Niemann, H. ; Arnold, J. ; Sagerer, G.
Author_Institution :
Lehrstuhl fuer Inf., Univ. Erlangen-Nurnberg, West Germany
Abstract :
The accuracy of optical flow computation is investigated using gradient methods and global optimization. It considered the influence of systematic and statistical errors on the constraint equation, and of systematic and statistical errors on the iterative solution of the optimization of the continuity constraint. Approaches to error reduction are suggested. Experimental results are given using synthetic images with exactly known properties demonstrating the efficiency of error reduction techniques
Keywords :
optimisation; pattern recognition; picture processing; accuracy; constraint equation; continuity constraint; error reduction; global optimization; gradient methods; iterative solution; optical flow computation; pattern recognition; picture processing; statistical errors; systematic errors; Constraint optimization; Differential equations; Fluid flow measurement; Gradient methods; Image motion analysis; Layout; Optical computing; Optimization methods; Velocity measurement;
Conference_Titel :
Pattern Recognition, 1988., 9th International Conference on
Conference_Location :
Rome
Print_ISBN :
0-8186-0878-1
DOI :
10.1109/ICPR.1988.28450
