Paper Contents
Abstract
Real-time recognition and tracking of objects can be of great significance in the area of surveillance, traffic analysis, and cars without drivers. This paper proposes a strongly lightweight and scalable system with YOLOv8, used as object tracker and ByteTrack, to achieve multi-object tracking. It is implemented in Python and libraries, including Ultralytics, OpenCV, Torch, and Supervision. It can accept webcam and video file input, generate labeled annotated frames denoting class and tracker id and track identity through time. It enables the application of the system in real-time; 25 30 FPS are achieved on standard hardware, proving the efficiency of this system. It is the focus of this study to note that detection and tracking of smart and responsive environments with deep learning is a powerful combination.
Copyright
Copyright © 2025 pinki rani. This is an open access article distributed under the Creative Commons Attribution License.
