Unbiased Estimation of Ellipses by Bootstrapping

Javier Cabrera(1) and Peter Meer(2)

(1)Department of Statistics
(2)Department of Electrical and Computer Engineering
Rutgers University, Piscataway, NJ 08855, USA

A general method for eliminating the bias of non-linear estimators using bootstrap is presented. Instead of the traditional mean bias we consider the definition of bias based on the median. The method is applied to the problem of fitting ellipse segments to noisy data. No assumption beyond being independent identically distributed (i.i.d.) is made about the error distribution and experiments with both synthetic and real data prove the effectiveness of the technique.

Appeared in IEEE Trans. Pattern Anal. Machine Intell, 18, 752-756, 1996.
Return to Research: Bootstrap as a tool for computer vision       Return to List of Publications
Download the paper