A Streamlit WebRTC web app that can identify whether a face comes from a real person or a fake person and prevents the system from giving false verification.

https://github.com/jomariya23156/face-recognition-with-liveness-web-login

I would highly suggest all of you to take a look at this wonderful repo which served as the inspiration for my project.

The drawback of the approach taken by jomariya23156 is that although he claims that it is a web app it cannot be used in a server-client scenario as it lacks basic webrtc features.

For more info, please check out this link: https://blog.streamlit.io/how-to-build-the-streamlit-webrtc-component/

• calculates real and fake ratios
• uses Streamlit's WebRTC features

Start developing locally.

Fork the repository. then clone the repo locally by doing -

git clone https://github.com/birdowl21/Face-Liveness-Detection-Anti-Spoofing-Web-App.git

pip install virtualenv
python -m venv [env-name]
[env-name]\Scripts\activate 

cd Face-Liveness-Detection-Anti-Spoofing-Web-App

pip install -r requirements.txt

streamlit run app.py

You should now have the application running and accessible at http://localhost:8501.

You could deploy the app to cloud platforms such as Streamlit-sharing and Heroku.

Deployment links:

• https://blog.jcharistech.com/2019/10/24/how-to-deploy-your-streamlit-apps-to-heroku/#:~:text=%20How%20to%20Deploy%20Your%20Streamlit%20Apps%20to,CLI.%20%20...%20The%20yourappname%20is...%20More%20
• https://towardsdatascience.com/deploying-a-basic-streamlit-app-ceadae286fd0#:~:text=To%20deploy%20the%20app%20to%20Streamlit%20Sharing%2C%20we,the%20screenshot%20of%20the%20deployment%20page%20on%20streamlit.io.

If you are facing issues with deploying the app remotely, please refer to this link: https://docs.streamlit.io/knowledge-base/deploy/remote-start

After cloning & setting up the local project you can push the changes to your github fork and make a pull request.

git add .
git commit -m "feat: added new stuff"
git push YOUR_REPO_URL develop

• can run on only 3-4 devices at a time.
• doesn't work well in bright background light.
• performance varies from browser to browser: Works fast on Chrome and Firefox but is slow on Edge.