- Download MaskedFace-Net dataset from https://github.com/cabani/MaskedFace-Net
- Move the content to get the following directory structure:
├── mask_detector
│ ├── data <- All your data modules should be located here!
│ │ ├── dataset <- Whole dataset
│ │ │ ├── mask <- Part of the dataset containing properly masked faces
│ │ │ │ ├── 00000 <- Dataset chunk containing images
│ │ │ │ │ ├── 00001.jpg <- Image
│ │ │ │ │ ├── 00002.jpg
│ │ │ │ │ └── ...
│ │ │ │ ├── 01000
│ │ │ │ └── ...
│ │ │ └── no_mask <- Part of the dataset containing incorrectly masked faces
│ │ │ ├── 00000 <- Dataset chunk containing images
│ │ │ │ ├── 00001.jpg <- Image
│ │ │ │ ├── 00002.jpg
│ │ │ │ └── ...
│ │ │ ├── 01000
│ │ │ └── ...
│ │ └── dataframe <- Place to store generated dataframes
│ └── ...
└── ...
First, install dependencies
# clone project
git clone https://github.com/HTD-Health/ai-smart-mirror.git
# install project
cd ai-smart-mirror
pip install -r requirements.txt
Next, navigate to any file and run it.
# module folder
cd mask_detector
# run module
python mask_classifier.py
We are using Dataplicity
for remote access to Raspberry Pi.
If you want to have access to Raspberry Pi for test purposes send your email address and request to be added to Jędrzej Polaczek.
Repository is under path home/pi/Workplace/ai-smart-mirror
To make changes (like changing branch) you need to run the command as superuser
.
To make it be you need to type su pi
(and execute this command) and after that enter the appropriate password.
Quick guide: https://docs.dataplicity.com/docs/superuser