A Real-Time Face Tracking System Based On A Single PTZ Camera

Abstract

It is evident that the effectiveness of education is strengthened from the assistance of distance learning which provides lectures online. The benefits of online education include enabling more distance courses based on local programs and increasing participation and interaction between the students and the instructor. Nowadays, conventional systems commonly implement a combination of static and PTZ cameras. However, such systems are not only costly but also require operators and high computational power in exchange.

Thus, this thesis proposes a real-time face tracking system based on a single PTZ camera as a cost effective solution by minimizing hardware requirements and functioning automatically. The proposed system focuses on the delay possible to occur due to the movement of the PTZ camera and the network delay which varies the video frame rate which alters the performance from a software perspective. The main contributions include the low cost and flexibility regarding installation. Preliminaries are introduced as a basis of the proposed system such that hardware is maintained to be minimal and universal while software is retained to use less computational power. The proposed system minimizes the delays to maintain pace with the subject of interest, provides a smooth and natural movement of the camera as if an actual operator controls the camera, and produces competitive results regarding performance compared to conventional systems.