lightweight-face-detection-using-Dlib-and-python

small API for using Dlib in python to create face detection and facial landmark detection system. achieve real-time performance on edge devices and CPU with a high tolerance for posed faces.

How to use

clone project, copy files to your project and you are good to go. example :

1- import wrapper class

from detector_light_dlib import FaceDetectorLightDlib

2- create an instance and load the facial landmark and face detection models

detection_system = FaceDetectorLightDlib()
detection_system.load()

3- call your detector on images or video frames

bounding_boxes, facial_points = detection_system.detect(image)

4- clean your models for saving space (it's very tiny but still matter for some applications!)

detection_system.clean()

dependencies

openCV dlib numpy

expected output

a NumPy array of bounding boxes for faces and NumPy array of n element each contains x,y coordinates of 5 facial points (two for each eye and one for the nose)

have fun !!