The aim of my project is to create a face detection and recognition app for check-in and check-out at a workplace. This project has three parts, the first one is face detection – yolo3, second part is face recognition or verification to be more precise – VGG-face descriptor, and the last part is the GUI that assembly the whole application.

I stated this project from scratch, so I used the exiting code and model to speed up my process.

Yolo3:

The code and everything about yolo3 which I used in my project was learnt from Mr. Rokas Balsys. He shows us how to download and label our desire object’s dataset as well as how to train the model.

This is the link to his tutorial: https://pylessons.com/YOLOv3-custom-data/

His tutorial is very detail and very helpful.

VGG-face:

There are many options like Facenet and OpenFace, but I chose VGG-face.

I also learn get the code and modified the code from Mr. Sefik Ilkin Serengil, his blog contains a detail explanation for the code as well as the source code for you to download. I copy the code and modified it to suit my project.

https://sefiks.com/2018/08/06/deep-face-recognition-with-keras/ for more details.

https://www.youtube.com/watch?v=WORpfhg1FbU

RTX 2060 | i5-9400F | 16Gb RAM DDR4 | tensorflow-GPU 2.0 | Keras 2.2.4 | python 3.6.1