https://github.com/smaranjitghose/sightguardai
Capitalizing moondream's capabilities to build a CCTV frame-on-framer analyzer
https://github.com/smaranjitghose/sightguardai
computer-vision llm localllm moondream python security streamlit vlm
Last synced: 3 months ago
JSON representation
Capitalizing moondream's capabilities to build a CCTV frame-on-framer analyzer
- Host: GitHub
- URL: https://github.com/smaranjitghose/sightguardai
- Owner: smaranjitghose
- License: mit
- Created: 2025-01-11T20:51:42.000Z (4 months ago)
- Default Branch: master
- Last Pushed: 2025-01-11T20:55:26.000Z (4 months ago)
- Last Synced: 2025-01-11T21:34:45.241Z (4 months ago)
- Topics: computer-vision, llm, localllm, moondream, python, security, streamlit, vlm
- Language: Python
- Homepage:
- Size: 0 Bytes
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.MD
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
Awesome Lists containing this project
README
# 👁️ SightGuardAI
An intelligent surveillance analysis tool that automatically detects and describes key events in video footage using Moondream Vision AI.
## 🌟 Features
* 🎬 Smart frame extraction from surveillance videos
* 🤖 AI-powered scene description generation
* 🎯 Intelligent key frame detection using similarity analysis
* 📊 Comprehensive data visualization with interactive tables
* ⌚ Precise timestamp tracking
* 📈 Frame-by-frame similarity comparison
* 🔄 Support for multiple video formats (MP4, AVI, MOV)
* 📱 Clean, responsive web interface
## 🖼️ Demo
## 🔧 Prerequisites
* Python 3.11 or higher
* Web Browser
* Moondream API key from [Moondream Console](https://console.moondream.ai) or download the model file from [here](https://docs.moondream.ai/specifications)
## 📥 Installation
1. Clone the repository:
```bash
git clone https://github.com/smaranjitghose/sightguardai.git
cd sightguardai
```
2. Create and activate virtual environment:
```bash
# Windows
python -m venv env
.\env\Scripts\activate
# Linux/Mac
python3 -m venv env
source env/bin/activate
```
3. Install required packages:
```bash
pip install streamlit moondream python-dotenv pillow opencv-python scikit-learn pandas
```
## 💡 Usage
1. Launch the application:
```bash
streamlit run app.py
```
2. Access the web interface:
```
http://localhost:8501
```
3. Enter your Moondream API key in the sidebar
4. Upload a surveillance video file
5. Click "Analyze" to begin processing
6. View results in the interactive table and key frames grid
## 🛠️ Troubleshooting
1. **Memory Issues**
* Adjust frame extraction interval for longer videos
* Ensure sufficient system RAM
* Close other memory-intensive applications
2. **API Errors**
* Verify API key validity
* Check internet connection
* Confirm API usage limits
3. **Video Processing**
* Ensure video format compatibility
* Check file corruption
* Verify file permissions
## 🤝 Contributing
Contributions are welcome! Please follow these steps:
1. Fork the project
2. Create your feature branch (`git checkout -b feature/AmazingFeature`)
3. Commit your changes (`git commit -m 'Add some AmazingFeature'`)
4. Push to the branch (`git push origin feature/AmazingFeature`)
5. Open a Pull Request
## 📝 License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
---
Made with ❤️ by [Smaranjit Ghose](https://github.com/smaranjitghose)