Process camera feed for head pose estimation is a Python application for computer vision live face parameterization
Using dlib's face landmark predictor, I added my implementation of a
real-time by building a graphics pipeline to support the 2D to 3D head
pose estimation method built by Satya Mallick in the referenced code
below.
models/ contains the models used in this example we use Facial Landmark detection 68 points.
*one must download shape_detector_68_facial_landmarks.dat because it is too large a file to host here.
media/ contains images and video.
Installation
Open a terminal in the headpose directory and run (with sudo if needed on your system):
pip install -r requirements.txt
Now you should have installed the necessary packages
You still need to download dlib model file: "shape_predictor_68_face_landmarks.dat"