Fast detection and tracking of faces in uncontrolled environments for autonomous robots using the CNN-UM

Abstract

ABSTRACT: We present a real-time system for the detection and tracking of faces in uncontrolled environments suitable for autonomous robots. The system is based on a CNN algorithm using a CNN-UM chip connected to a robot, which had to perform a face tracking task using a web camera mounted on a pan-tilt zoom unit. The experimental evaluation of the system demonstrates fast, reliable tracking of faces under uncontrolled lighting conditions and with temporary occlusions of the tracked face. 1.