PEDESTRIAN RECOGNITION BY USING KERNEL DESCRIPTORS.

Recognition of people in images is important for many applications in computer vision. This paper presents an experimental study on pedestrian classification. We investigate the recently developed kernelbased features in order to represent an image and two learning algorithms: the popular Support Vector Machine (SVM) and Genetic Programming (GP). Numerical experiments are performed on a benchmark dataset consisting of pedestrian and non-pedestrian (labeled) images captured in outdoor urban environments and indicate that the evolutionary classifier is able to perform better over SVM. [ABSTRACT FROM AUTHOR]