LUCID AI is a browser-based fatigue detection system that uses real-time webcam input and facial landmark tracking to monitor eye closure and detect possible drowsiness.
Built for accessibility and instant use, the application runs directly in the browser with no installation or backend required.
https://sleep-detector-web.vercel.app/
Try it instantly in your browser.
Fatigue-related accidents often begin in quiet moments — a delayed reaction, a prolonged blink, or a momentary lapse in attention.
I first explored this problem through a local Python/OpenCV prototype, then rebuilt it as a modern web application so anyone could access it instantly without installing software.
Previous Prototype:
https://github.com/AdityaSharma-Git3207/Sleep-Detector-Program
- Real-time webcam monitoring
- Eye closure detection using facial landmarks
- Instant fatigue alerts
- Fully browser-based experience
- No backend required
- Responsive UI dashboard
- Public live deployment
- HTML5
- CSS3
- JavaScript
- MediaPipe Face Mesh
- WebRTC
- Vercel
- User grants webcam permission
- Face landmarks are tracked in real time
- Eye openness is continuously measured
- If prolonged closure is detected, an alert is triggered
- Clone the repository:
git clone https://github.com/AdityaSharma-Git3207/Sleep-Detector-Web.git
cd Sleep-Detector-WebAditya Sharma
Engineering Student • Builder • Open Source Contributor