This repository is used for developing a face recognition system, even when you are wearing a face mask.

1. Using python3.7 is recommend, or you can create a virtual env with python version 3.7

2. Run this command to install packages

pip install -r requirements.txt

3. Download our pretrained model on GDrive then put them into models folder like models/top_patch.h5 and models/face_mask_detection.pb

1. Data preparation

For the first time, if you want to add your own images data, just prepare and add the data following the format

input_image/
      - giang/
            + mask/
                  ~ mask_img1
                  ~ mask_img2
                  ~ mask_img3
            + no_mask/
                  ~ no_mask_img1
                  ~ no_mask_img2
                  ~ no_mask_img3
      - linh/
            + mask/
                  ~ mask_img1
                  ~ mask_img2
                  ~ mask_img3
            + no_mask/
                  ~ no_mask_img1
                  ~ no_mask_img2
                  ~ no_mask_img3
      ...

Please create your own folder following the given format.

2. Create database Run the following command to create embeddings saved inside database folder

python create_embeddings.py

3. Run camera.py to activate recognition process using your camera

python camera.py