Getting started

1. Download

Donwload necessary packages from ftp.

Download If you are going to use GPU for computing

Ubuntu 18.04

• Ubuntu 18.04 CPU binary package
• Ubuntu 18.04 CPU dev package
• Ubuntu 18.04 GPU binary package
• Ubuntu 18.04 GPU dev package

NN's algorithms. Choose needed version

• Face Detector 200
• Face Detector 201
• Face Detector 210
• Face Detector 211
• Alignment 103
• Builder 103
• Builder 104
• Builder 104
• Builder 105
• Builder 106
• Builder 200
• Builder 210
• Builder 211
• Age&Gender 100
• Builder 211
• Quality 100
• Face Recognition 100

Centos 7

• Centos 7 CPU binary package
• Centos 7 CPU dev package
• Centos 7 GPU binary package
• Centos 7 GPU dev package

NN's algorithms. Choose needed version

• Face Detector 200
• Face Detector 201
• Face Detector 210
• Face Detector 211
• Alignment 103
• Builder 103
• Builder 104
• Builder 105
• Builder 106
• Builder 200
• Builder 210
• Builder 211
• Age&Gender 100
• Quality 100
• Face Recognition 100

Windows

All libraries, executable files, headers, docs and algorithms are in the archive.

• Windows x64 CPU
• Windows x64 GPU

Android

All libraries, executable files, headers, docs and algorithms are in the archive.

• Android armv7
• Android armv8

iOS

All libraries, executable files, headers, docs and algorithms are in the archive.

• iOS 13 armv8

2. Install

Linux-only: Correct work of this function requires the executable file, which have read access to the attribute product_uuid udev device dmi/id (usually it's /sys/devices/virtual/dmi/id/product_uuid file).

Ubuntu 18.03

• install downloaded packages with

  dpkg -i <package_name1>
  dpkg -i <package_name2>
  ...
  dpkg -i <package_nameN>
  apt-get install -f

Centos 7

• install downloaded packages with

  yum -y install epel-release 
  yum -y localinstall <rpm_package1>
  yum -y localinstall <rpm_package1>
  ...
  yum -y localinstall <rpm_packageN>

for GPU builds

• Latest NVIDIA drivers
• NVIDIA Cuda 10.1 Update 1
• NVIDIA Cudnn v7.6.1 (June 24, 2019), for CUDA 10.1

Windows

• Unpack archive into preferred directory.
• Add FACE_SDK_BIN_ROOT with value face_sdk_install_dir%/share into system environment
• Install Visual C++ Redistributable for Visual Studio 2015 x64
• for GPU builds:
  • Latest NVIDIA drivers
  • NVIDIA Cuda 10.1 Update 1
  • NVIDIA Cudnn v7.6.1 (June 24, 2019), for CUDA 10.1

Android

Android Studio or see manuals how to integrate native lib for your IDE.

iOS

See manuals how to integrate native lib for your IDE.

3. Get the license

How to get the license

4. Try face_sdk_utils

face_sdk_utils guide

5. Learn docs and see examples

Docs

Examples