Note: This project won first place at SRCHacks 2018, it has been originally posted on HackerEarth.
We saw that there was a lot of trash and litter in the streets and thought that we could find a way to clean the streets all the while making it more fun and interactive.
What it does
sWEep is a platform that enables communities to quickly and effectively clean their streets through the power of computer vision, the Internet, and crowdsourcing. Every user of sWEep has the ability to take a picture of trash they see in their community, which is then classified as a certain type of trash by our machine learning algorithm and entered into the database. At that point, a pinpoint will appear on all users' maps showing this new piece of trash with information about how many points its worth and map directions, encouraging them to help clean it up as soon as possible. Signing into the sWEep app using your Google account gives us connections with all of your Google contacts, allowing you to compete with your friends to make the world a better place!
How we built it
We built it using React Native, and Expo. We used Firebase for Authentication using Google Sign In and the Database. We used Clarifai for the Machine Learning algorithm/model.
Challenges we ran into
We had never used React Native before, although most of us had used React in the past so it wasn't too bad. Linking our code to Firebase was especially challenging, we had a lot of bugs with placing markers on Google Maps.
Accomplishments that we're proud of
We're proud of making it function almost perfectly and have it do everything that we planned for it to do. It works on both iOS and Android using the Expo app.
What we learned
We learned a lot about React and React Native.
What's next for sWEep
Ideally, we'd rewrite the entire project with better code and SDKs so that the app runs better, and does not require the Expo app to run (as a standalone app or something).