This document proposes using computer vision and machine learning to monitor exams for cheating. Cameras would track student behavior and feed data to a model during exams. The model analyzes posture, head/eye position, lips, hands and individual behaviors to identify potential cheating in real-time or offline. Alerts are generated and a report shows if behaviors broke rules. The solution aims to address issues like impersonation and mass cheating due to lack of proper monitoring in remote areas and few invigilators.