Computer Vision · Case study
AI Cheating Detector
A real-time exam-integrity system for Arab Open University that uses computer vision to flag cheating during online exams the instant it happens.
- When
- May 2026
- Stack
- Python · FastAPI · YOLOv8 · OpenCV · MediaPipe · Telegram API
The challenge
Remote and on-site exams are easy to cheat on, and manual invigilation doesn't scale — one person can't reliably watch dozens of live video feeds at once.
What I built
- YOLOv8 + face recognition detect phones, head-turns, and unknown people in frame
- Multi-violation detection running in real time over a WebSocket live-video feed with on-screen overlays
- Thread-based frame processing to keep detection fast under load
- Instant Telegram alerts so invigilators get an evidence-backed flag the moment something happens
Highlights
- Multi-violation detection (phone, head turn, unknown person)
- WebSocket live video with overlays
- Thread-based frame processing
The outcome
Invigilators get instant, specific alerts with visual evidence instead of trying to watch everything at once — turning proctoring from guesswork into a focused response.
Want something like AI Cheating Detector?
Message me on WhatsApp or see all services.