Comparative Study of Vehicle Tracking Techniques

This review offers a comprehensive examination of vehicle detection, estimation, and advanced classification using the latest deep learning methodologies. Specifically, it conducts a performance evaluation of three prominent methods: YOLOv8, ByteTrack, and Supervision. YOLOv8 is renowned for its rapid recognition capabilities, delivering swift and precise vehicle classification. ByteTrack introduces a novel approach to tracking through temporal stability, ensuring dependable traffic monitoring across frames. On the other hand, Supervision em- ploys hierarchical supervision to enhance feature representation and classification accuracy, particularly beneficial in scenarios involving obstructed objects. The research aims to provide insights into the strengths and limitations of each approach in complex real-world scenarios. It evaluates their efficacy in accurately detecting vehicles, tracking them amidst dense foliage, and precisely tallying their occurrences through rigorous testing and analysis. This study contributes to understanding whether these methods are suitable for various applications such as urban planning, traffic management, and surveillance.