Skip to content

Latest commit

 

History

History
125 lines (91 loc) · 7.61 KB

README.md

File metadata and controls

125 lines (91 loc) · 7.61 KB

This application is used to check everything is ok and running as fast as expected. It's open source and doesn't require registration or license key.

More information about the benchmark rules at https://www.doubango.org/SDKs/face-liveness/docs/Benchmark.html.

Dependencies

The SDK is developed in C++11 and you'll need glibc 2.27+ on Linux and Microsoft Visual C++ 2015 Redistributable(x64) - 14.0.24123 (any later version is ok) on Windows. You most likely already have these dependencies on you machine as almost every program require it.

If you're planning to use OpenVINO, then you'll need Intel C++ Compiler Redistributable (choose newest). Please note that OpenVINO is packaged in the SDK as plugin and loaded (dlopen) at runtime. The engine will fail to load the plugin if Intel C++ Compiler Redistributable is missing on your machine but the program will work as expected with Tensorflow as fallback. We highly recommend using OpenVINO to speedup the inference time. See benchmark numbers with/without OpenVINO at https://www.doubango.org/SDKs/face-liveness/docs/Benchmark.html#core-i7-windows.

Debugging missing dependencies

To check if all dependencies are present:

GPGPU acceleration

  • On x86-64, GPGPU acceleration is disabled by default. Check here for more information on how to enable it.

Peformance numbers

These performance numbers are obtained using version 0.5.0. You can use any later version.

We ran the benchmark application for #100 times (loops) on 720p (1280x720) images. The number of milliseconds reported is for #100 frames which means you have to divide by 100 to get the time it takes to process a single frame. For example, 2140 millis in the below table means it takes 21 milliseconds to process a single frame, in terms of frame rate it means we can process 46 frames per second.

Spoof (Parallel) Spoof (Sequential) Disguise (P) Disguise (S)
AMD Ryzen 7 3700X 8-Core
RTX 3060
Ubuntu 20		 2140 millis
46.70 fps		 2977 millis
33.58 fps		 2062 millis
48.48 fps		 4130 millis
24.20 fps
Intel(R) Xeon(R) E3-1230 v6 @ 3.50GHz
GTX 1070
Ubuntu 18		 2655 millis
37.65 fps		 3553 millis
28.13 fps		 2721 millis
36.73 fps		 4436 millis
22.54 fps
Intel(R) i7-4790K @4.40GHz
No GPU
Windows 8		 7197 millis
13.89 fps		 7281 millis
13.73 fps		 11480 millis
8.71 fps		 11643 millis
8.58 fps

Some important notes:

  • You can increase the speed (significantly) by choosing higher value for detect_face_minsize JSON config entry (default is 128).
  • The engine is faster on spoofs. That's normal as there is more checks on disguises and genuines.
  • Parallel mode is faster than sequential mode on GPU. When parallel mode is enabled we perform detection and liveness check in //.

The test image looks like this: Test image

Pre-built binaries

If you don't want to build this sample by yourself, then use the pre-built versions:

On Windows, the easiest way to try this sample is to navigate to binaries/windows/x86_64 and run binaries/windows/x86_64/benchmark.bat. You can edit these files to use your own images and configuration options.

Building

You'll need CMake to build this sample.

  • Create build folder and move into it: mkdir build && cd build

To generate the build files:

  • Windows (Visual Studio files): cmake .. -DCMAKE_GENERATOR_PLATFORM=x64 -DCMAKE_BUILD_TYPE=Release
  • Linux (Makefile): cmake .. -G"Unix Makefiles" -DCMAKE_BUILD_TYPE=Release

To build the project:

  • Windows: Open the VS solution and build the projet
  • Linux: Run make to build the project

Testing

After building the application you can test it on your local machine.

Usage

Benchmark is a command line application with the following usage:

benchmark \
      --image <path-to-image-to-process> \
      --assets <path-to-assets-folder> \
      [--tokenfile <path-to-license-token-file>] \
      [--tokendata <base64-license-token-data>]

Options surrounded with [] are optional.

  • --image Path to an image (JPEG/PNG/BMP) to process. This image will be used to evaluate the liveness detector. You can use default image at ../../../assets/images/disguise.jpg.
  • --assets Path to the assets folder containing the configuration files and models.
  • --tokenfile Path to the file containing the base64 license token if you have one. If not provided, then the application will act like a trial version. Default: null.
  • --tokendata Base64 license token if you have one. If not provided, then the application will act like a trial version. Default: null.

Examples

  • On Linux x86_64, you may use the next command:
LD_LIBRARY_PATH=../../../binaries/linux/x86_64:$LD_LIBRARY_PATH ./benchmark \
    --image ../../../assets/images/disguise.jpg \
    --assets ../../../assets \
    --loops 100 \
    --parallel true
  • On Windows x86_64, you may use the next command:
benchmark.exe ^
    --image ../../../assets/images/disguise.jpg ^
    --assets ../../../assets ^
    --loops 100 ^
    --parallel true