In-Class Examinations Proctoring System Using YOLO Architecture

This research paper is a proof of concept using the YOLO architecture to develop an in-class examination monitoring system. It focuses on detecting common cheating behaviors like head movements and using forbidden objects. The dataset that is utilized to illustrate each contains a large range of cheating behaviors' examples, while considering the patterns of students who are not engaging in cheating at a moment. The study outlines a structured work plan to systematically achieve robust detection and tracking capabilities for identifying instances of academic dishonesty during examinations.