Real-Time People Counting System using YOLOv8 Object Detection
Abrar Elaoua, Mohamed Nadour, Lakhmissi Cherroun, Abdelfattah Elasri
Abstract
This paper presents a comprehensive real-time people counting system that utilizes the advanced YOLOv8 object detection algorithm. The proposed method aims to accurately track individuals within a video stream and provide precise counts of people entering and exiting specific areas of interest. The system combines state-of-the-art computer vision techniques, leveraging the robust object detection capabilities of YOLOv8, along with efficient tracking mechanisms and region-based analysis. The system demonstrates exceptional accuracy and robustness in people-counting tasks through extensive experimental evaluations conducted across various scenarios. The results highlight the system's effectiveness in crowd management, occupancy analysis, and surveillance applications.