Fix your hood

• 

  leaderboards

• 

  picture function

• 

  solve a problem

• 

  map overview

Inspiration

The power of street maps lies in their liability to guide us. We want to implement a game that fixes differences in the digital representation of the real world.

What it does

Our app lets people update Open Street Map Issues in a gamified way. People take pictures that are classified by our visual recognition model while incentivized in a competitive environment.

How we built it

Frontend and Backend are separated, making working on it together easier. We used flutter for the mobile app implementation, Django Rest Framework to connect to our Database and design our API, and Microsoft Azure for the ML model.

Challenges we ran into

We built two more models (CNNs) with transfer learning. Setting the hyperparameters right was a challenge. The frontend was the other challenge. We never worked with the Flutter Framework before. Learning a new framework while building our fully working prototype under time pressure was quite the challenge.

Accomplishments that we're proud of

Having a working prototype never having worked on an app before.

What we learned

Building apps with Flutter and implementation of Neural Networks.

What's next for the project

Visualize players' impact by drawing conquest zones around the issues they fixed and finalizing User authentication by adding the corresponding API Calls to the existing authentication backend logic.

Built With

• azure
• django
• flutter
• python
• tensorflow