Detecting Cheating vs Normal Behavior in Classroom Settings with YOLOv11 Pose Estimation
🚀 AI in Action: Detecting Cheating vs Normal Behavior in Classroom Settings with YOLOv11 Pose Estimation 🚀
I’m thrilled to share my latest project — a custom-trained YOLOv11 pose estimation model built to automatically detect and classify actions as “normal” or “cheating” from video footage.
🎯 What the system does
✅ Processes any video source — recorded or live feed
✅ Detects actions in real time with high accuracy
✅ Overlays pose skeletons and class labels for each person
✅ Generates a fully annotated video for review and analysis
💡 Why it matters
This approach can be adapted beyond classrooms — from exam monitoring to workplace compliance and behavioral analytics — delivering both detection accuracy and visual transparency through pose keypoints.
🛠 Tech Stack
YOLOv11 Pose Estimation (Ultralytics)
Python + OpenCV for video processing
Custom dataset labeled for normal vs cheating actions
🔍 Next Steps
I’m working on real-time deployment with live alerts for proactive monitoring.