Welcome to FaceRecLib’s documentation!

The FaceRecLib is an open source tool that is designed to run comparable and reproducible face recognition experiments. To design a face recognition experiment, one has to choose:

  • an image databases and its according protocol,
  • face detection and image preprocessing algorithms,
  • the type of features to extract from the face,
  • the face recognition algorithm to employ, and
  • the way to evaluate the results

For any of these parts, several different types are implemented in the FaceRecLib, and basically any combination of the five parts can be executed. For each type, several meta-parameters can be tested. This results in a nearly infinite amount of possible face recognition experiments that can be run using the current setup. But it is also possible to use your own database, preprocessing, feature type, or face recognition algorithm and test this against the baseline algorithms implemented in the FaceRecLib.

For example, we created wrapper classes for the CSU Face Recognition Resources, which can be found in the xfacereclib.extension.CSU satellite package, including installation instructions for the CSU toolkit.

If you are interested, please continue reading:

Reference Manual

ToDo-List

This documentation is still under development. Here is a list of things that needs to be done:

Todo

Document the remaining parameters of the PLDA

(The original entry is located in /mgunther/test_checkouts/facereclib/doc/experiments.rst, line 339.)

Todo

Document the remaining parameters of the UBMGMM tool

(The original entry is located in /mgunther/test_checkouts/facereclib/doc/experiments.rst, line 380.)

Todo

Document the remaining parameters of the ISV tool

(The original entry is located in /mgunther/test_checkouts/facereclib/doc/experiments.rst, line 389.)

Todo

Document the JFA tool

(The original entry is located in /mgunther/test_checkouts/facereclib/doc/experiments.rst, line 399.)

Indices and tables