System requirements

Supported operating systems

Operating system	Architecture	Recommended version
Windows	x86-32	10
	x86-64	10
Linux	x86-32	Ubuntu 18.04
	x86-64	Ubuntu 18.04
	arm-32	Ubuntu 18.04
	arm-64	jetpack 4.4
Android	arm-32	api 21 - 30
	arm-64	api 23 - 30
iOS	arm-64	12 - 17.1

Available APIs

• C++ API (for Windows, Linux, Android, iOS)
• Java wrapper (for Windows, Linux, Android)
• C# wrapper (for Windows, Linux)
• Python wrapper (for Windows, Linux)
• Swift API (for iOS)
• Flutter plugin (for iOS, Android)

Hardware recommendations

The following hardware configuration recommended to achieve best performance*.

Desktop PCs (x86 64-bit CPU):

• CPU Intel Core i7-7700K @ 4.20GHz (4 Cores);
• 8 GB RAM;

* Based on the result of test on the FullHD (1920x1080) video stream with no more than 5 faces in the frame using method10v100_recognizer.xml recognition method.

Android (ARM 64-bit CPU):

• CPU Qualcomm Snapdragon 821 @ 2.15 GHz (4 Cores);
• 4 GB RAM;

* Based on the result of test on the VGA (640x480) video stream with 1 face in the frame using method10v30_recognizer.xml / method9v30mask_recognizer.xml recognition methods.

GPU acceleration

GPU acceleration is available at:

• Windows x86 64-bit
• Linux x86 64-bit
• NVIDIA Jetson (JetPack 4.3/4.4)
• Android (Legacy API only)

Windows/Linux

• General requirements:

  • Software requirements:

    Windows:

    • Nvidia GPU Driver >= 441.22
    • Microsoft Visual C++ Redistributable for Visual Studio 2019

    Linux:

    • Nvidia GPU Driver >= 440.33

  • Hardware requirements:

    • CUDA compatible GPU (NVIDIA GTX 1050 Ti or better)

• For CUDA 11:

  • Software requirements:

    Windows:

    • CUDA Toolkit 11.0 ≤ version < 12
    • cuDNN v8.8.0 - v9.0.0 for CUDA 11.x

    Linux:

    • CUDA Toolkit 11.0 ≤ version < 12
    • cuDNN v8.8.0 for CUDA 11.x

  • Hardware requirements:

    • CUDA compatible GPU (NVIDIA GTX 1050 Ti or better up to and including RTX 3090)

• For CUDA 10:

  • Software requirements:

    Windows:

    • CUDA Toolkit 10.1
    • cuDNN v7.6.5 for CUDA 10.1

    Linux:

    • CUDA Toolkit 10.1
    • cuDNN v7.6.5 for CUDA 10.1

  • Hardware requirements:

    • CUDA compatible GPU (NVIDIA GTX 1050 Ti or better, but below the 30xx versions)

You can also use pre-built docker containers with CUDA support, such as nvidia/cuda:10.0-cudnn7-devel-ubuntu16.04 (note that some licenses can be unavailable in this case).

NVIDIA Jetson

• System requirements:
  • JetPack 4.3 or 4.4*

* Tests were performed on the Jetson TX2 and Jetson NX modules.

You can select jetson_jetpack_4.3_4.4 component in Face SDK installation wizard. By default, it uses the build for jetpack 4.4. If a build for jetpack 4.3 is required, move all files from the lib/jetpack-4.3 directory to the lib directory.