With 3 million+ cases of COVID-19 in North America, social distancing needs to be maintained more than ever. Moreover, as cities start to open up due to economical pressures, a means to know which parts of the city are crowded to maintain social distancing measures is needed. That's where our application comes in.
What it does
We used facial recognition and google maps business activity data to create a heatmap of hotspots around a city or area so people know which places they can avoid going through.
How we built it
We used many video streams processed in “real-time” with OpenCV and fed through object detection in Python. We also scraped data from business on Google Maps. The data was then aggregated, prepared, and pushed to Firebase Database with real-time updates. The data was then pulled from the Realtime Database and displayed in a heatmap on Android.
Challenges we ran into
It was hard figuring out how to link the database to the heatmap but once we figured it out it turned out great.
Accomplishments that we're proud of
We were able to complete a project without the benefit of a whiteboard. It was also the first time many of us had used Android Studio and Firebase.
What we learned
We learned to work with strangers and become friends through the experience. Additionally, we learned that it's always okay to ask for help when needed. In terms of technical skills, we improved our skills in front-end android app development as well as working with Google's Firebase for the backend. Moreover, we learned how to use facial recognition to detect crowds and translate that into a heatmap on Google Maps.
What's next for Distance Better
After completing our proof of concept, we aim to expand our database to include other provinces in Canada as well as US states so more people can be benefitted from social distancing. After the pandemic, we aim to pivot into an app to help people avoid lines at crowded shopping malls, restaurants and events.