BLM511-FaceRecognition

Edit variables in config.py file under ./face-recognition/ directory.

Setup

Before setup make sure that you have correct version of python. If you have installed different version of python on your system you may want to use pyenv for install any specific version.

There are two suggested ways to run this application. In case you choose the solution with pipenv setup then it will install required python automatically if pyenv is available.

Setup and run with pipenv package

Learn more details about pipenv

Install pipenv

pip install pipenv

Install dependencies

pipenv install

Run the application

pipenv run python face-recognition

Setup and run with pip

Learn more details about pyenv

Create virtual environment

python3 -m venv ./.venv

Activate virtual environment

source ./.venv/bin/activate

Install dependencies

pip install -r requirements.txt

Run the application

python face-recognition

Deactivate virtual environment

deactivate

Description of Data and Source

The extended Yale Face Database B contains 16128 images of 28 human subjects under 9 poses and 64 illumination conditions. The data format of this database is the same as the Yale Face Database B.

Georghiades, A., Belhumeur, P., & Kriegman, D. (2001). From few to many: Illumination cone models for face recognition under variable lighting and pose. IEEE Trans. Pattern Anal. Mach. Intelligence, 23(6), 643–660.