Skip to main content
Version: 3.24.0 (latest)

Face detection and cropping, image quality assessment, gender and age estimation

The bin directory contains the executable file test_facecut:

demo​

The program demonstrates tracking, detection and cropping of faces, determination of anthropometric points and angles, as well as assessment of face quality, determination of gender and age, emotions, whether the face belongs to a living person (analysis of a color image from a camera).

To quickly launch the sample with default parameters, go to the bin folder and double-click the left mouse button to launch the demo.exe program.

Location of C++ demo

On the left side of the program window there are Face SDK components that can be enabled/disabled by left-clicking:

  • rectangles – face bounding rectangle
  • angles – head rotation angles
  • quality – image quality
  • liveness – whether the face belongs to a real person
  • age and gender – age and gender
  • base cut, full cut, token cut – face cropping type (base, full frontal TIL, conditional frontal TIL)
  • points – anthropometric points
  • face quality – face image quality
  • angles vectors – angle vectors
  • emotions – emotions
  • masked face – detecting the presence of a mask on the face
  • eyes openness – checking whether the eyes are closed or open

Result of running demo.exe

Error messages are displayed in the console when they occur.

You can also run the demo sample by specifying certain parameters (for example, if you have an online license). Launch parameters:

  • capturer_conf – path to the detector configuration file (for more information on configuration file types, see Capturer Class)
  • config_dir – path to the conf/facerec directory
  • dll_path – path to the library file libfacerec.so or facerec.dll
  • license_dir – path to the directory with the license; specify if you need to change the default license directory

Source code: demo.cpp

The OpenCV library is required for building.