Conturbà

Inspiration

Noticing the obvious severity of the coronavirus pandemic, we decided to create a simulation for people to play around with and visualize the spread among a crowd, along with a live mask and social distancing detection system, to help institutions in containing the spread. Unfortunately, the live mask and social distancing systems proved too difficult to incorporate into the simulation's parameters given the time-frame of the hackathon, and so we dropped this aspect of the project.

What it does

Our project does one thing. It randomly generates a map of infected/uninfected people and simulates how coronavirus spreads among that map.

How we built it

We built the simulation and user interface using ReactJS and deployed this on Vercel using NextJS.

Challenges we ran into

We had trouble with layout and styling in ReactJS and utilizing CSS Libraries
We had trouble implementing a form to input parameters for how to run the simulation.
We planned on implementing live face mask detection that would integrate into the simulation. However, the integration step failed.
We planned on gathering coordinates from the images/video feed to plot in the simulation. However, the integration between the two aspects of our project did not end up working.
Gathering percentages and calculating probabilities for the spread based on distance/mask-wearing

Accomplishments that we're proud of

We're proud of how the simulation worked out. Unfortunately, the integration step simply did not work, and we didn’t see any way of putting the mask/distance detection in the website.

What we learned

We learned a lot about React JS, especially layouts and styling, and some things about React in general. We also learned about how to apply machine learning to live video data, even though we didn't apply this aspect.

What's next for Conturbà

We intend to continue working on the project and to work on being able to implement the machine learning work that we tried working on. In the future, we also could send the live-feed detection algorithm to schools (or other crowds) to detect whether people are following social distancing guidelines. If we had just a little more time, we would be able to use the positions of people from videos to make better simulations of how coronavirus spreads in a real crowd .