Face Extractor

Extract faces from images inside a folder.

CLI utility to extract faces from an image or a folder. Mostly based on the awesome article by Adrian Rosebrock.

Installation

OS X & Linux:

Clone the repository and run:

git clone https://github.com/rickaa/face_extract.git
chmod +x face_extract.py

Install dependecies if needed:

pip install numpy
pip install opencv-contrib-python

If you use Pipenv for virtual environments management and have OpenCV installed with HomeBrew you can do:

ln -s "$(brew --prefix)"/lib/python3.7/site-packages/cv2*.so "$(pipenv --venv)"/lib/python3.7/site-packages

If you want to run the test_face.py file:

pip install matplotlib

Usage example

Place the images you want to extract faces from in the input folder

./face_extract.py -f ./input/ -p deploy.prototxt.txt -m res10_300x300_ssd_iter_140000.caffemodel

Detect faces in image test1.jpg but without extracting them

./face_extract.py -i ./input/test1.jpg -p deploy.prototxt.txt -m res10_300x300_ssd_iter_140000.caffemodel

Inner workings

The file test_face.pyis meant to be run with a kernel so that you can see step by step what is going on in a clearer way than just looking at the for loop in the main script. In order to work with it place a file called test.jpg in the input folder.

For example:

Release History

0.1.0
- The first proper release

Contributing

Fork it (https://github.com/rickaa/face_extract/fork)
Create your feature branch (git checkout -b feature/fooBar)
Commit your changes (git commit -am 'Add some fooBar')
Push to the branch (git push origin feature/fooBar)
Create a new Pull Request