Web app demo for Pytorch Retinaface

This app uses the pytorch retinaface library to detect faces from either a video or an image. I forked ternaus retinaface_demo repo and added the video functionality to the web application.

Install dependencies

pip install -r requirements.txt

Run App from terminal

1. clone this repo

git clone https://github.com/phemiloyeAI/face-detection.git

2. change the directory to the path containing the streamlit app

cd ./app/app.py

3. Run app.py script

streamlit run app.py

Deploy App

If you wish to deploy this app yourself, here is how to:

• sign up to open a streamlit cloud account.
• Connect your streamlit account to your github account containing this repo you cloned.
• On your streamlit cloud account, create an app. You can call it any name.
• Click on "Paste GitHub URL".
• Paste the link to your github repo in the bar.
• And finally click on "Deploy", right below the bar.

The code for the network: https://github.com/ternaus/retinaface \

The Web app: https://share.streamlit.io/phemiloyeai/face-detection/app/app.py