Eyedentify is a web application that uses computer vision to detect and analyze eye strain. It integrates with Roboflow’s object detection model to track eye strain levels and notifies users when excessive strain is detected.
• Live camera feed for real-time eye detection
• Roboflow model integration for eye strain analysis
• Screen time tracking
• Auth0 authentication for user login/logout
• Notifications for excessive eye strain
1️⃣ Clone the Repository
git clone https://github.com/your-username/eyedentify.git
cd eyedentify
2️⃣ Install Dependencies
npm install
3️⃣ Set Up Environment Variables
Create a .env file in the root directory and add:
REACT_APP_ROBOFLOW_API_KEY=your_api_key_here
REACT_APP_MODEL_PUBLIC_KEY=your_model_key_here
REACT_APP_MODEL_VERSION=your_model_version_here
4️⃣ Run the App
npm start
Then, open http://localhost:3000/ in your browser.
1. Log in using Auth0 authentication.
2. Click Start Model to enable the camera and begin eye detection.
3. The app processes video frames using Roboflow’s API and detects eye strain.
4. If prolonged strain is detected, the app notifies the user.
• React.js – Frontend framework
• Roboflow API – Computer vision model
• Auth0 – Authentication
• Inference.js – Runs Roboflow models in the browser