FaCiPa: FACIAL PARALYSIS DIAGNOSIS ALGORITHM (Facial Landmarks 68 coordinates)

Running the Application

The versions used in the PyCharm IDE to run Python codes are listed below.

Libraries and Packages to be Installed with Priority

👉 Pip Package Manager

👉 Python 3.5 Version

👉 Flask for API if needed

---

All Libraries and Packages in the Project

✔️ Python Version:3.5

✔️ pip:19.1.1

✔️ flask:1.1.1

✔️ cmake:3.15.3

✔️ dlib:19.4.0 || latest version( 19.18.0 )

✔️ opencv-python:4.1.1.26

✔️ imutils:0.5.3

In the project, facial paralysis algorithm is run based on 68 coordinate models. The code file that contains the algorithm and the shape-predictor model is facial_landmarks.py, and the Flask API code written for the mobile platform of the project is located in the api_trying.py file.

If only the paralysis algorithm will be run, the photo to be detected on the imageP variable based on the file must be specified manually.

> python facial_landmarks.py --shape-predictor shape_predictor_68_face_landmarks.dat \--image face.jpg

If server-based work is to be done, it is sufficient to run the api_trying file.

> python api_trying.py

Setting Pip

Setting Ubuntu 16.04 PIP

https://www.rosehosting.com/blog/how-to-install-pip-on-ubuntu-16-04/

A view from the project: