Our vision is to enhance people living standard and to transform the traditional industry through our face recognition technology. Our technology can be use by developers as a tool to easily build applications based on our technology.
Face recognition technology can help our life more secure, efficiency and convenience. There are many ways to adapt face recognition technology.

Ayonix Face-ID SDK finds human faces from Digital Media (Image/Video) and recognizes them if there is a record of the person. Ayonix Face-ID SDK is a cutting edge facial recognition software which is especially designed to be integrated into existing applications or new applications.

Ayonix Face-ID SDK covers enrollment, verification, and identification. It runs multiple algorithms for finding of human faces and eyes in images, for tracking faces in video streams and for analyzing and checking face characteristics to support the total face recognition solution.

General Description

The Ayonix Software development kit provides our customers the flexibility to integrate biometric security into your own application, or existing network and hardware security solutions. Fully customizable and easy to implement, these solutions are built upon our most powerful algorithms to secure every point of action


1Ayonix Face-ID Face detection
Given an image, find faces. Ayofa face detection technology returns the cropped face coordinates in the image. Minimum face detection criteria is 20x20pixel(or 15 pixels between the eyes)
2Ayonix Face-ID Identification
Creates face templates (AFID), and recognizes faces against templates (one-to-many matching). Minimum face size is limited to 64x64pixels (or 25 pixels between the eyes)
3Demo Applications
An Example of face detection, enrollment and Identification is implemented in C, C#. Sample sources is fully available to our customers.

SDK Features

  • Speedy gender, age, smile analytics
  • Powered 3D engine
  • Multiple face detection(Up to 20 people)
  • Face pose support(30 degree)
  • Stable recognition in Backlighted environments
  • 99.6% accuracy on Feret-DB
  • 100ms fast matching(World's fastest engine)

What SDK Include

  • Powerful learning-based facial detection
  • 3D-based grid face alignment
  • Local face feature extraction
  • Improved illumination correction
  • State-of-the-art landmark localization on images
  • Advanced face feature classification

SDK Versions

Windows Linux   Android
OS Windows 7/8 Windows 2008/2012 server Kernel2.6 Android 4.4.2
Bit 32bit/64bit 32bit/64bit 64bit
CPU Intel Atom/Celeron i3/i5/i7/Xeon Intel Atom/Celeron i3/i5/i7/Xeon ARM
Template 2KB 2KB 2KB
Samples C/C++/C# C/C++ Java
License Dongle/Software Dongle/Software Software