Inspiration
Our motivation to join HackHarvard this weekend was to make an impact around us. One thing that our team members have in common is the love for our furry animals, so we tackled on a project to save our furry friends around us. When our motive and our passion united, it synthesized to the project we built this weekend: Pawz - for a second and save an animal.
What it does
Pawz targets two audiences: a user who's lost their furry friend, and a user who's reporting furry friend sightings in the neighbourhood. Both users must fill out a furry friend information form, where they type in information such as the furry friend's animal type (only cat or dogs are supported by the program at the moment), gender, fur color, eye color, an image of the furry friend and the location last found. Our app then uses algorithms to find similarities between the lost furry friend and furry friends who have been sighted nearby, and alert the owner if there has recently been a similar furry friend found nearby.
How we built it
We decided to build an app to allow users to keep track in a timely manner; it just takes too much time to take a photo, run to a nearby desktop and fill out the information. Thus, we built the app using Android Studio, with support from Google Maps API and Firebase for storing the furry friends' data.
Challenges we ran into
One of the biggest challenge we had to resolve was to streamline and minimize the resources required to track down our furry friends. We wanted to track down as much information as possible but by thinking from the end user's perspective, we shortened the list down to be as compact and as useful as possible.
What we learned
I think if we still had more time, we could improve our matching algorithm and also incorporate deep-learning mechanism if possible, to better identify, locate and match our furry friends.
What's next for Pawz
- more furry friend support
- improved functionality for matching and identification
- notification system which will allow user to be notified immediately if a high percentage / viable match shows up nearby
Log in or sign up for Devpost to join the conversation.