README.md

Face Recognition SDK for Windows and Linux - Fully On Premise

Overview

The world's 1st Completely Free and Open Source Face Recognition SDK from Faceplugin for developers to integrate face recognition capabilities into applications. Supports real-time, high-accuracy face recognition with deep learning models.
This is on-premise face recognition SDK which means everything is processed in your phone and NO data leaves the device
You can use this SDK on Windows and Linux

Please contact us if you need the SDK with higher accuracy.

Key Features

• Real-Time Face Recognition: Can detect and recognize faces from live video streams. Currently only supports face recognition from an image.
• High Accuracy: Built with deep learning models trained on large datasets.
• Cross-Platform: Compatible with Windows and Linux.
• Flexible Integration: Easy-to-use APIs for seamless integration into any project.
• Scalable: Works on local devices, cloud, or embedded systems.
• Python SDK: Comprehensive support for Python with extensive documentation and examples.

Applications

This Face Recognition SDK is ideal for a wide range of applications, including:

• Time Attendance Systems: Monitor arrivals and depatures using face recognition.
• Security Systems: Access control and surveillance.
• User Authentication: Biometric login and multi-factor authentication.
• Smart Devices: Integration into IoT devices for smart home or office applications.
• Augmented Reality: Enhance AR applications with real-time facial recognition.
• Retail: Personalized marketing and customer analytics.

Installation

Please download anaconda on your computer and install it. We used Windows machine without GPU for testing.

create anaconda environment

conda create -n facesdk python=3.9

activate env

conda activate facesdk

install dependencies

pip install -r requirements.txt

compare 1.jpg and 2.png in the test directory.

python run.py

APIs and Parameters

GetImageInfo(image, faceMaxCount): returns face bounding boxes, landmarks and feature embeddings
get_similarity(feat1, feat2): returns similarity between two feature embeddings. 0 to 100
Threshold: value to determine if two embeddings belong to same person, default = 75

List of our Products

• FaceRecognition-LivenessDetection-Android
• FaceRecognition-LivenessDetection-iOS
• FaceRecognition-LivenessDetection-Javascript
• FaceLivenessDetection-Android
• FaceLivenessDetection-iOS
• FaceLivenessDetection-Linux
• FaceRecognition-LivenessDetection-React
• FaceRecognition-LivenessDetection-Vue
• Face Recognition SDK
• Liveness Detection SDK
• ID Card Recognition

Contact