<p>Recently, the research on person identification through video analysis has gained more attention in the fields like security, and surveillance. The conventional models developed for video processing and person identification often face challenges like increased latency, lack of accuracy, and large computational requirements. In recent years, Edge Computing (EC) technology has emerged as a promising solution for addressing these issues. In this study, a novel Hybrid Deep Learning (DL)-based EC framework was proposed for precise detection of persons by analyzing videos. The developed algorithm integrates the effectiveness of Recurrent Neural Network with Ant Lion Optimization (RNN-ALO) for identifying the persons. The RNN in the proposed algorithm analyzes the spatial and temporal features within each video frame, while the ALO fitness determines the center point of the bounding box to identify the human as a single point. In addition, a heat map was utilized for locating accurate bounding box locations. The presented methodology was executed in Python tool and validated using the CEPDOF and Human detection databases. The implementation outcomes manifest that the presented algorithm obtained 99.75% accuracy, 99.86% recall, and 99.36% precision. Furthermore, a comparative assessment was made with the conventional models and it validated that the proposed strategy earned improved results.</p>