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.
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()
openCV dlib numpy
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 !!