🚨 An AI-powered real-time violence detection and alert system designed to enhance public safety
🚀 Quick Start • 📖 Documentation • 🤝 Contributing • ⭐ Give us a star!
eyeview.ai leverages cutting-edge YOLO deep learning models integrated with Flask backend and React frontend to create a comprehensive safety monitoring solution.
| Feature | Description | Status |
|---|---|---|
| 🔍 Real-time Detection | YOLO-powered violence detection with lag-free processing | ✅ Active |
| 📹 Evidence Recording | Automatic clip recording and Firebase storage | ✅ Active |
| 📱 Instant Alerts | SMS/Call notifications via Twilio integration | ✅ Active |
| 📊 Live Dashboard | Real-time monitoring and incident management | ✅ Active |
| 📱 Mobile Support | Compatible with mobile camera feeds | ✅ Active |
| 🏙️ Smart City Ready | Optimized for public safety and surveillance | ✅ Active |
🎯 Detection → 📹 Recording → ☁️ Storage → 📱 Alert → 🚨 Response
- 📹 Evidence Capture: System records and stores short clips in Firebase
- ⚡ Instant Alerts: SMS/Calls sent via Twilio to authorities
- 📊 Dashboard Monitoring: Real-time feed monitoring and incident history
- 🔍 Review System: Comprehensive incident management interface
The project uses a YOLOv11 nano model (yolo11n.pt) that was fine-tuned for violence detection. The training ran for 28 epochs, with the best model saved at epoch 28 based on the highest mAP50-95 score of 0.53977.
📦 eyeview.ai
├── 🐍 Backend/ # Flask backend (YOLO + Twilio + Firebase)
│ ├── 🎯 eye-view.py
│ ├── 📋 requirements.txt
│ ├── 🔧 config/
│ └── 📂 models/
│
├── ⚛️ EyeView-frontend/ # React frontend (Vite + Tailwind)
│ ├── 📁 public/
│ │ ├── 🖼️ assets/
│ │ └── 📄 index.html
│ ├── 📁 src/
│ │ ├── 🧩 components/
│ │ ├── 📄 pages/
│ │ ├── 🎨 styles/
│ │ └── ⚙️ utils/
│ ├── 🔒 .env
│ ├── 📦 package.json
│ ├── ⚡ vite.config.js
│ └── 🎨 tailwind.config.js
│
├── 📖 README.md
└── 📄 LICENSE# 📞 Twilio Integration (SMS/Call Alerts)
TWILIO_ACCOUNT_SID=your_twilio_account_sid
TWILIO_AUTH_TOKEN=your_twilio_auth_token
TWILIO_PHONE_NUMBER=+1234567890 # 📱 Twilio registered number
ADMIN_PHONE_NUMBER=+919876543210 # 🚨 Alert recipient number
# 🔥 Firebase Configuration
FIREBASE_CONFIG=your_firebase_config_json
FIREBASE_STORAGE_BUCKET=your_storage_bucket# 🔥 Firebase Web SDK Configuration
REACT_APP_FIREBASE_API_KEY=your_firebase_api_key
REACT_APP_FIREBASE_AUTH_DOMAIN=your_project.firebaseapp.com
REACT_APP_FIREBASE_PROJECT_ID=your_project_id
REACT_APP_FIREBASE_STORAGE_BUCKET=your_project.appspot.com
REACT_APP_FIREBASE_MESSAGING_SENDER_ID=your_sender_id
REACT_APP_FIREBASE_APP_ID=your_app_id
# 🌐 API Configuration
REACT_APP_API_BASE_URL=http://localhost:5000git clone https://github.com/VyasaAi/eyeview.ai.git
cd eyeview.ai# Navigate to backend directory
cd Backend
# Create virtual environment
python -m venv venv🔄 Activate Virtual Environment:
💻 Windows (PowerShell)
.\venv\Scripts\activate🐧 macOS/Linux
source venv/bin/activate📦 Install Dependencies & Run:
# Install required packages
pip install -r requirements.txt
# Start the backend server
python eye-view.py🟢 Backend Status: Server running on
http://localhost:5000
# Navigate to frontend directory
cd EyeView-frontend
# Install dependencies
npm install
# Start development server
npm run dev🟢 Frontend Status: Dashboard available at
http://localhost:5173
Open your browser and navigate to the displayed URL (typically http://localhost:5173) to view the EyeView dashboard.
🌟 EyeView thrives on community contributions! Join us in making public spaces safer.
-
🍴 Fork the Repository
# Fork on GitHub, then clone your fork git clone https://github.com/<your-username>/eyeview.ai.git cd eyeview.ai
-
🌿 Create Feature Branch
git checkout -b feature/awesome-feature-name
-
💻 Make Your Changes
- Test both frontend and backend thoroughly
- Follow existing code conventions
- Add comments and documentation
-
📝 Commit Your Work
git add . git commit -m "feat: add awesome new feature"
-
🚀 Push & Create PR
git push origin feature/awesome-feature-name
Then create a Pull Request on GitHub!
🎯 Code Standards
- Follow PEP 8 for Python code
- Use ESLint and Prettier for JavaScript/React
- Write clear commit messages using conventional commits
- Include tests for new features
- Update documentation as needed
🧪 Testing Requirements
- Test new features locally before submitting
- Ensure existing functionality remains intact
- Add unit tests for critical functions
- Test on multiple browsers/devices when applicable
📚 Documentation
- Add inline comments for complex logic
- Update README for new features
- Include usage examples
- Document API changes
| Type | Description | Examples |
|---|---|---|
| 🐛 Bug Fixes | Fix existing issues | Memory leaks, UI bugs, API errors |
| ✨ New Features | Add functionality | New detection models, dashboard widgets |
| 📚 Documentation | Improve docs | Setup guides, API documentation |
| 🎨 UI/UX | Design improvements | Better dashboard, mobile responsiveness |
| ⚡ Performance | Speed & optimization | Faster detection, reduced latency |
| 🧪 Testing | Add test coverage | Unit tests, integration tests |
- 🤖 Advanced AI Models: Integration with more sophisticated detection algorithms
- 🌐 Multi-language Support: Internationalization for global deployment
- 📊 Analytics Dashboard: Detailed reporting and insights
- 🔐 Enhanced Security: Advanced authentication and encryption
- 📱 Mobile App: Native mobile applications for field operations
- 🏗️ Scalability: Kubernetes deployment and microservices architecture
This project is licensed under the GPLv3 - see the LICENSE file for details.
- YOLO Team for the incredible object detection framework
- Open Source Community for continuous support and contributions
- Safety Advocates who inspire us to build better security solutions
📬 Questions? Issues? Ideas?
💬 Start a Discussion • 🐛 Report Bug • 💡 Request Feature
Built with ❤️ for a safer world