Anotace:
A design of the second stage of classifier based on the recently proposed trainable similarity measure, which is specifically designed for the supervised classification of images.The proposed measure is based on local matches in a set of regions within an image which increases its robustness.The configuration of local regions is derived specifically for each prototype by a training procedure. The comparison to the state-of-art AdaBoost classifiers on a real-world pedestrian recognition problem is presented.