This repo uses a Face Detector caffe-model trained using jointly several databases: IJB-A, FDDB and WIDER.

In order to setup the folders and download the caffemodel, please run first the setup.sh

One the setup has been made, the file you are looking for is get_faces.py. There are two modules you want to use, for instance:

from get_faces import __init__, face_from_file
net=__init__()
face = face_from_file('/path/to/your/img', net)

• Caffe
• OpenCV