Recent research aims to enhance the accuracy and efficiency of Person Re-identification (Re-ID) systems by developing advanced deep learning models, with a particular focus on Convolutional Neural Networks (CNNs). These systems automate the detection and recognition of human faces in surveillance footage, reducing manual monitoring and improving cost-effectiveness. By leveraging CNNs for feature extraction and similarity measurement, they enable robust identification of individuals across scenarios, facilitating tasks like finding missing persons and tracking suspects. Advances in tracking algorithms utilizing deep learning-based object detection and association techniques further enhance the system’s capabilities in accurately re- identifying individuals and tracking their movements across different camera views. However, challenges such as variations in lighting and appearance persist, which the proposed research aims to address through the introduction of two key models: a Face Recognition library and a CNN employing efficient accuracy metrics. This shift toward video- based systems and utilization of real-world surveillance datasets seeks to provide a comprehensive solution for Person Re-identification and path tracking, contributing to enhanced security and operational efficiency in domains like law enforcement and institutional administration.