🎭 Face Recognition API

A FastAPI-based face recognition system that can identify faces in photos and videos. The system maintains a database of known people and can recognize them in real-time.

🚀 Features

• Add People: Register new people with their photos
• Face Recognition: Identify people in uploaded photos
• Video Processing: Recognize faces in video files
• People Management: View, manage, and delete registered people
• REST API: Full RESTful API with Swagger documentation
• Web Interface: User-friendly HTML interface for testing

📋 Requirements

• Python 3.8+
• OpenCV
• face-recognition library
• FastAPI
• See requirements.txt for full list

🛠️ Installation

1. Clone the repository:

git clone <repository-url>
cd face-recognition-api

2. Create a virtual environment:

python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

pip install -r requirements.txt

🏃‍♂️ Running the Application

1. Navigate to the API directory:

cd API/src

2. Start the server:

python main.py

The server will start on http://localhost:8000

🖥️ Using the Web Interface

Open test_page.html in your web browser to access the user interface.

Available Functions:

1. Add Person: Upload a photo with a person's name to register them
2. Recognize Photo: Upload a photo to identify people in it
3. Recognize Video: Upload a video file to identify people throughout the video
4. View Database: See all registered people
5. Statistics: View system statistics

📡 API Endpoints

Database Management

• POST /api/person/add - Add new person
• GET /api/person/list - List all people
• GET /api/person/{id} - Get specific person info
• DELETE /api/person/{id} - Delete person

Recognition

• POST /api/person/recognize - Recognize faces in photo
• POST /api/video/recognize - Recognize faces in video
• POST /api/person/recognize-base64 - Recognize faces in base64 image

System

• GET / - API health check
• GET /api/stats - System statistics

📚 API Documentation

When the server is running, visit:

• Swagger UI: http://localhost:8000/docs
• ReDoc: http://localhost:8000/redoc

📁 Project Structure

API/
├── src/
│   └── main.py          # Main application file
├── uploads/             # Uploaded photos storage
├── persons_data.csv     # Database file
├── test_page.html       # Web interface
├── debug_video.html     # Debug interface for video
└── requirements.txt     # Python dependencies

🎥 Video Processing

When processing videos:

• Frame Interval: Process every Nth frame (default: 15)
  • Lower values = more accurate but slower
  • Higher values = faster but may miss short appearances
• Key Frames: Up to 10 annotated frames are returned
• Results: Includes timestamps and confidence scores

🔧 Configuration

Key settings in main.py:

• CSV_FILE: Database file location
• UPLOAD_DIR: Directory for uploaded files
• frame_interval: Video processing interval

🐛 Troubleshooting

Common Issues:

1. "No module named 'face_recognition'"

   • Make sure you've installed all requirements
   • On some systems, you may need to install dlib first

2. Video processing is slow

   • Increase the frame_interval parameter
   • Use smaller video files
   • Ensure your system has adequate resources

3. Face not detected

   • Ensure good lighting in photos
   • Face should be clearly visible and frontal
   • Photo resolution should be adequate

📝 Data Storage

• Database: CSV file storing person info and face encodings
• Photos: Stored in the uploads directory
• Temporary: Video files are deleted after processing

🔐 Security Notes

• CORS is configured to accept all origins (*) - restrict this in production
• No authentication is implemented - add security layers for production use
• Uploaded files are stored locally - implement proper file management

🤝 Contributing

Feel free to submit issues, fork the repository, and create pull requests for any improvements.

📄 License

This project is open source. Please check the license file for more details.