M. Borga, H. Knutsson and T. Landelius,
Learning Canonical Correlations,
In The 10th Scandinavian Conference on Image Analysis (SCIA'97),
Lappeenranta, Finland, pp. 1-8, June 1997.
M. Borga, H. Knutsson and T. Landelius,
Learning Canonical Correlations,
In The 10th Scandinavian Conference on Image Analysis (SCIA'97),
Lappeenranta, Finland, pp. 1-8, June 1997.This paper presents a novel learning algorithm that finds the linear combination of one set of multidimensional variates that is the best predictor, and at the same time finds the linear combination of another set which is the most predictable. This relation is known as the canonical correlation and has the property of being invariant with respect to affine transformations of the two sets of variates. The algorithm successively finds all the canonical correlations beginning with the largest one. It is shown that canonical correlations can be used in computer vision to find feature detectors by giving examples of the desired features. When used on the pixel level, the method finds quadrature filters and when used on a higher level, the method finds combinations of filter output that are less sensitive to noise compared to vector averaging.