Face Recognition System

Overview

This is a comprehensive face recognition system built with Django and various AI technologies. The system provides robust face detection, feature extraction, and face matching capabilities through both RESTful APIs and a web interface.

Features

• Face Library Management: Register, update, query, and delete face information
• Face Detection: Detect faces in images and extract the largest face by default
• Feature Extraction: Extract 512-dimensional face feature vectors
• 1:1 Face Verification: Compare two faces for similarity
• 1:N Face Identification: Match a face against the entire database
• Similar Face Search: Find similar faces in the database (top-5 results)
• Web Management Interface: Visual interface for face library management

System Requirements

• Python 3.7+
• GPU support (for optimal performance)
• CUDA compatible environment (for GPU acceleration)

Installation

1. Clone the repository

git clone https://github.com/senthree3/faceRecognition.git
cd faceRecognition

2. Set up a virtual environment

python -m venv venv
# On Windows
venv\Scripts\activate
# On Unix or MacOS
source venv/bin/activate

3. Install dependencies

pip install -r requirements.txt

4. Install PyTorch (uncomment the PyTorch packages in requirements.txt or install manually)

# For CUDA 11.x
pip install torch==2.0.0+cu117 torchvision==0.15.0+cu117 torchaudio==2.0.0 --index-url https://download.pytorch.org/whl/cu117

5.Model Installation

# Download the AI models package
# Download the required AI models file (aiModels.zip) from:
wegt https://drive.google.com/file/d/1w2eQzqurwTwB-5yoonRFGXb4YbHZCscc/view?usp=sharing

# Extract the model files
# Extract the downloaded aiModels.zip file to the faceRecognition directory
unzip aiModels.zip -d /path/to/project/faceRecognition/

6. Set up Django

python manage.py migrate
python manage.py createsuperuser

7. Run the development server

python manage.py runserver

Configuration

Configure the Django settings in faceRecognition/settings.py to match your environment. You may need to adjust database settings, static files, and security settings.

Web Management Interface

The system includes a web interface for face library management. Access it through:

http://{IP}:{PORT}/admin/

API Documentation

The system provides RESTful APIs for face recognition operations. All API endpoints use standardized request and response formats with encryption and authentication mechanisms.

Base URL

http://{IP}:{PORT}/ai/openAbility/v1/

Authentication

All API requests require authentication via:

• access_key: Access key identifier
• secret_key: Secret key for encryption and signing
• time_stamp: Unix timestamp (milliseconds)
• request_id: UUID4 string
• sign: MD5 signature
• data: AES-encrypted request data

• access_key and secret_key obtain from Web Management

API Endpoints

Face Library Management

1. Register Face: /faceReg
2. Delete Face: /faceRegDelete
3. Update Face: /faceReg (with existing face_id)
4. Query Face by ID: /faceRegIDQuery
5. Query Faces with Pagination: /faceRegQuery

Face Recognition

1. Face Detection: /faceDet
2. Feature Extraction: /faceFeatureExtraction
3. 1:1 Face Verification: /faceVerify
4. 1:1 Face Verification Against DB: /faceVerifyDB
5. 1:N Face Identification: /face1N
6. Similar Face Search: /faceSimilarRetrieval

Example Usage

For detailed API usage examples, parameter specifications, and response formats, please refer to the API Documentation included in the project.

Security Considerations

• Keep your access_key and secret_key confidential
• Use HTTPS in production
• Implement proper authentication and authorization
• Consider data privacy regulations when storing face data

Contact Information

For technical support or inquiries:

• Email-1: senthree30@gmail.com
• Email-2: 717192305@qq.com
• GitHub Issues: https://github.com/senthree3/faceRecognition/issues

License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.

Acknowledgements

• This project uses various libraries including Django, OpenCV, FAISS, ONNX Runtime, and PyTorch
• Thanks to all contributors and the open-source community