Inspiration
In today’s digital world, screen time has skyrocketed — from remote work to online learning, most of us spend hours glued to screens without realizing the toll it takes on our eyes. Millions suffer from digital eye strain, blurred vision, headaches, and reduced productivity, yet few tools offer real-time, intelligent intervention.
We were inspired to build GazeIntel to bridge this gap — combining AI-powered gaze detection with predictive health insights, all packaged into an accessible Chrome extension. Our mission is simple: 🔹 Help users build healthier screen habits 🔹 Prevent digital fatigue before it happens 🔹 Promote long-term eye wellness through smart technology
What it does
GazeIntel is an innovative Chrome Extension and desktop solution designed to safeguard your vision during extended screen usage. Powered by advanced AI and computer vision technology, GazeIntel uses your device’s webcam to intelligently track eye movement and screen engagement. It provides real-time alerts, predictive eye strain warnings, and personalized break reminders—promoting healthier digital habits and reducing long-term eye fatigue.
How we built it
🔹We built GazeIntel as a lightweight Chrome Extension and desktop-integrated solution using the following technologies:
🔹Frontend (Extension UI): HTML, CSS, JavaScript for seamless browser integration and real-time interaction
🔹AI & Eye Detection: TensorFlow.js for real-time face and eye landmark detection using the webcam
🔹Prediction Engine: Custom-trained lightweight ML model to analyze blink patterns and gaze stability for detecting early signs of eye strain
🔹Break Notification System: JavaScript timers and event triggers for delivering personalized reminders based on activity duration and fatigue risk
🔹Backend (Optional Desktop Companion): Python with OpenCV for advanced gaze tracking and extended capabilities (if webcam access is limited via browser)
🔹Data Handling: Local storage APIs to track daily usage trends and generate user-friendly visual insights
🔹We focused on privacy-first design by ensuring all camera processing happens locally in the browser — no frames are stored or sent externally. Our goal was to deliver a smart, accessible, and secure eye wellness solution that anyone can use without complex setup.
Challenges we ran into
🔹Real-time eye tracking in the browser was difficult due to performance and accuracy limitations.
🔹Lighting and webcam quality varied across devices, affecting detection reliability.
🔹Ensuring user privacy while running all AI models locally was technically demanding.
🔹Designing non-intrusive break reminders that users find helpful required careful tuning.
🔹Faced cross-browser compatibility issues during Chrome Extension deployment.
Accomplishments that we're proud of
🔹Built the core of our Chrome Extension with real-time eye tracking.
🔹Integrated AI models locally to ensure user privacy.
🔹Set up a basic alert system for eye strain prevention.
🔹Currently refining detection accuracy and UI experience.
🔹We’re still actively working on enhancements and fixing challenges as we go.
What we learned
🔹Gained hands-on experience with real-time eye tracking using TensorFlow.js in the browser.
🔹Learned how to optimize AI models for performance without compromising privacy.
🔹Understood the importance of user-friendly alerts and subtle UI design in wellness apps.
🔹Discovered the challenges of cross-browser compatibility and webcam variability.
🔹Most importantly, we learned how to collaborate under time pressure and keep improving, even when the product is still in progress.
What's next for GazeIntel
🔹Improve UI/UX for better user engagement and smoother alerts.
🔹Add customizable break schedules and health tips based on usage patterns.
🔹Explore integration with desktop apps and mobile platforms for wider accessibility.
🔹Enhance eye strain prediction accuracy with deeper ML model tuning.
🔹Enhancing integration with React and cross-platform.
Log in or sign up for Devpost to join the conversation.