This paper focuses on exploring the applicability, strengths, and weaknesses of deep learning tracking algorithms to track students’ movement and the practical application of it in complex environments such as universities. We focus on three tracking algorithms: BoT-SORT which is recognized for its accuracy and complex matching methods; ByteTrack which is established for accurate and efficient tracking; and DeepSORT which incorporates appearance learning to improve on the traditional tracking algorithm. Then, with the help of real-life data, the effectiveness of each algorithm is discussed at the same time in terms of the given criteria. we used these algorithms to monitor and populate the number of students coming to the classes and exiting the classes. Therefore, the evaluation shows that ByteTrack and BoT-SORT are more accurate in preserving identities and performing better in different tracking scenarios compared to DeepSORT. The findings presented are useful in enhancing the physical security and organizational management of a university by incorporating modern technologies in tracking systems.