This is a program created to train a model in order to recognize some people on an image. This model will be trained from photos of these people, and after that it will be able to recognize any of these people on any image. This program uses Pytorch to create his neural network and python bindings of OpenCV (cv2) for image processing.
First of all, we should add images of heads of these people in the folder images/train. Each group of photos ( preferably the photos of head) of each person must be placed in a folder named after this person's name. After that, we must train our model by executing this command in the project root
python3 trainer.py
After that our model be trained, we must execute this command in the project root
python3 main.py path/to/image.jpg
Where path/to/image.jpg is the path to the image that we want to test.