Eye-Blink Detector
Overview
Real-Time Eye-Blink Anti-Spoof Detection System is a biometric liveness verification project built using advanced computer vision techniques. The FastAPI backend leverages Mediapipe Face Mesh and the Eye Aspect Ratio (EAR) algorithm to detect eye blinks and determine whether a face is live or spoofed through both REST APIs and WebSocket-based real-time streaming.
A blink state machine tracks recent blink patterns for reliable liveness detection. The Next.js frontend provides a responsive, camera-enabled interface that streams video frames to the backend and visualizes blink status, confidence, and counts in real time.
This project demonstrates practical anti-spoofing, biometric security, and real-time AI system deployment.
A blink state machine tracks recent blink patterns for reliable liveness detection. The Next.js frontend provides a responsive, camera-enabled interface that streams video frames to the backend and visualizes blink status, confidence, and counts in real time.
This project demonstrates practical anti-spoofing, biometric security, and real-time AI system deployment.
Technologies
- Nextjs
- Numpy
- OpenCV
- FastAPI
# Features
- Real-time eye-blink detection using computer vision
- Liveness verification to prevent spoofing attacks
- REST API and WebSocket support for flexible integration
- Responsive frontend with live video streaming
- Blink state machine for reliable detection