Sentinel is a computer vision project built to explore how AI can be used for real-time surveillance. It processes live video feeds, detects objects using a deep learning model, and can be extended to identify unusual events or security threats.
This project was developed to gain practical experience with computer vision, real-time inference, and integrating AI into real-world applications.
- Real-time object detection
- Live camera feed processing
- Fast inference using YOLO
- Modular project structure
- Easy to extend with new detection classes and features
- Python
- OpenCV
- YOLO
- FastAPI / Flask
- HTML, CSS, JavaScript
Clone the repository:
git clone https://github.com/yourusername/Sentinel.gitMove into the project directory:
cd SentinelInstall the required packages:
pip install -r requirements.txtRun the application:
python app.pySentinel/
├── backend/
├── frontend/
├── models/
├── dataset/
├── utils/
├── app.py
├── requirements.txt
└── README.md
Working on Sentinel helped me understand:
- Building applications with computer vision
- Using YOLO for object detection
- Processing video streams in real time
- Organizing AI projects for scalability
- Integrating machine learning into practical applications
Some improvements I plan to make include:
- Face recognition
- Multi-camera support
- Cloud deployment
- Real-time notifications
- Performance optimization
Dhruv Rangan
B.Tech Computer Science Engineering