ROBUST MULTIRESOLUTION ESTIMATION OF PARAMETRIC MOTION MODELS

This paper describes a method to estimate parametric motion models. Mo tivations for the use of such models are, on one hand, their efficienc y, which has been demonstrated in numerous contexts such as estimation , segmentation, tracking, and interpretation of motion, and on the oth er hand, their low computational cost compared to optical how estimati on. However, it is important to have the best accuracy for the estimat ed parameters, and to take into account the problem of multiple motion . We have therefore developed two robust estimators in a multiresoluti on framework. Numerical results support this approach, as validated by the use of these algorithms on complex sequences. (C) 1995 Academic P ress, Inc.