Object detection and localization system based on neural networks for Robo-Pong

This paper presents a vision system for a ping-pong player robot, called Robo-Pong. The robot employs color object detection techniques based on neural networks in its vision system. In this approach a quite simple architecture is employed to detect and localize objects in robotpsilas work space. The architecture is designed to be very easy-implement and also surprisingly fast to work on such a real-time system. Also a mapping system is attached to the object detection one, in order to estimate object locations. To increase the real-time in-field train capabilities of the system some early stopping methods were exploited to deal with such vast train data.