Inspiration
Petr stickers. Our dear mascot Peter reimagined into different forms of popular characters, and students running all around campus just to find them; a unique UCI culture. As Petr enthusiasts, we have all experienced running to the wrong location for a sticker drop, so we decided to make a tool that's able to guide Petr runners directly to where they need to be, for that shiny, fabulous sticker.
What it does
Collects an image from the user, and locates where in UCI that image belongs to, so Petr runners are able to run straight to the Petr drop without risk of going to the wrong place.
How we built it
For front end, we used React and javascript to make the interface of our website. For backend, we used Google Cloud Directions API to integrate Google Maps and Google Cloud Streetview API for image collection to train our computer vision algorithm, SIFT. Python was the main language for our backend.
Challenges we ran into
We had trouble coordinating work times because we all had other obligations over the Hackathon weekend, so communication during the first day was tough. We eventually delegated tasks on Saturday, but that cost us development time. Using Google Cloud API was also an obstacle because we weren't sure if we were going to exceed the free trial limit, in the end we didn't and was not billed. None of us were familiar with frontend, so frontend development took longer than expected, postponing the project's progress.
Accomplishments that we're proud of
We are super proud of each other for pulling through with the project despite our busy schedules. We are also glad that our project idea formation process was productive enough it gave us buffer time to deal with problems. Our frontend developers did a fantastic job given it was their first time learning React, and our backend group was able to produce completed and working code without any unfixable errors.
What we learned
We learned to better schedule our times and that it's possible reach a huge learning curve during a Hackathon if you have enough motivation.
What's next for PetrFindr
We intend to further develop PetrFindr into a mobile app, so it's more accessible to Petr runners who use instagram stories to get information about sticker drop spots.
Built With
- api
- computer-vision
- google-cloud
- html
- javascript
- python
- react
- sift
Log in or sign up for Devpost to join the conversation.