The main idea of this project is to learn hot to use YOLO to solve the object detection problem. More precisely, here I solve the problem of anti-spoofing: detecting a real person and detecting an attempt to fake his image. At the root you can find python scripts with cli, which can help you to store the data, order them in necessary order, train model and use it on inference.
Here my stack includes:
- YOLO via ultralytics;
- cli via click;
Here you can find a gif of me with model inference example:
This project based on awesome video by Murtaza Hassan.