A Real-time Automated System for Object Detection and Facial Recognition

Abstract

Object detection, facial recognition, and person identification are important tasks in computer vision with numerous real-life applications. The major goal of the proposed model is to identify people and recognize them in the images. In this paper, we propose a real-time automated system that combines power of both EfficientDet model for object detection and the FaceNet model for facial recognition to detect persons in an input image, recognize their faces, and label them with their corresponding names. The experimental study of the model takes place on COCO dataset and a custom dataset of images of students. This solution can be applied to various scenarios beyond education, such as in security and surveillance, healthcare, transportation, retail, and entertainment etc. The importance of the model lies in its ability to efficiently and accurately perform person identification and recognition in real-time scenarios, which can save time and resources and improve overall efficiency.

Keywords

• object detection
• facial recognition
• efficientdet facenet and coco (common objects in context)