HackZurich 2021

Team - Eye-Need-a-Break

The web app uses eye tracking to assess your level of distraction and shows the scores on a dashboard.

The eye gaze coordinates are extracted from the video data using the webgazer.js API. Only these coordinates are sent to the backend, protecting the user's privacy. On the Flask backend, the coordinates are stored in an SQLAlchemy database. A distraction score is calculated based on the standard deviation of these coordinates, relative to a random distribution over the whole screen. The time series and statistics of the score are shown on a dashboard powered by Plotly Dash.

Project Link : https://devpost.com/software/eye-need-a-break

To run the WebGazer-

1. Allow access to camera
2. On the popup click calibrate
3. Follow instructions in the next popup
4. Check console for live data points
5. To set a custom background click on browse
6. Once done you can click on save to download .csv files

Run backend and db

> pip install -r .\backend\requirements.txt
> python run_server.py

http://localhost:5000/apidocs/ - api description

Deploy

> sudo ufw allow 8050
> apt install git -y
> git clone https://github.com/HarishGuragol/eye-need-a-break.git
> cd eye-need-a-break
> apt install docker-compose -y
> docker-compose up

Useful commands

> docker rm -vf $(docker ps -a -q) // Delete all docker containers
> docker rmi -f $(docker images -a -q) // Delete all docker images

It may be useful before new deploy (delete old data)

Generate keys

openssl genrsa 2048 > host.key
chmod 400 host.key
openssl req -new -x509 -nodes -sha256 -days 365 -key host.key -out host.cert