Lokalizace chodců v dopravě pomocí obrazů

Abstract

The goal of this work is to get acquainted with methods for locating pedestrians which are already all around us today and help people improve their lives. In the future, such methods will be an integral part of everyday life. The work will mainly focus on specific methods, comparison of accuracy and time requirements among themselves. The implementation of detectors for the detection of individual pedestrians will be in different libraries for the Python language. The training of the models will be from a set of images from the ECP, Cityscapes and PRW datasets. Testing their functionality will subsequently be tested on sets of images of the ECP dataset and the Cityscapes dataset.