Face Recognition OnGoing¶
The goal of this package is to provide an “easy to reproduce” set of experiments in some large scale face recognition databases. This package is an extension of the bob.bio.base framework.
Installation¶
The installation instructions are based on conda (LINUX ONLY). Please install conda before continuing.
After everything installed do:
$ conda install bob.bio.face_ongoing
$ bob bio face_ongoing download_models <DESTINATION_PATH>
If you want to developt this package do
$ git clone https://gitlab.idiap.ch/bob/bob.bio.face_ongoing
$ cd bob.bio.face_ongoing
$ conda env create -f environment.yml
$ source activate bob.bio.face_ongoing # activate the environment
$ buildout
This software component contains all the necessary software stack to execute face recognition experiments, but doesn’t provide any data to test it. Click here to see how to prepare the database data before execute an experiment.
Face recognition baselines are available in the format of baseline via the command bellow:
$ bob bio baseline <BASELINE_NAME> <DATABASE_NAME>
You can use –help for more information:
$ bob bio baseline --help
To check it out the baselines and the databases available do:
$ resources.py --types baseline
$ resources.py --types database
Databases¶
This subsection describes the databases used in this work and some of the state-of-the-art results.
Baselines¶
This subsection presents the error rate results for each baseline/database.