Fauna Lookout

We decided to tackle USAA's challenge to help out Australia and the life that lives there. The destruction that is taking place in Australia is devastating and the need for new technology is ever more important. We wanted to help in any way we could and while we were searching we came across an estimate that over 500,000,000 animals have died from the fires, and we knew were to start. Fauna Lookout is a mobile app used for the safe finding, retrieval, and care of endangered and injured animals. Fauna Lookout uses googles firestore database for its ease of use and real time database. When a person who has the app comes across an animal they think is injured or in danger, they can open the app and drop a pin at their location alerting local officials so that they can find the animal and take any necessary action. Fauna Lookout also uses pythons keras library to train our model that recognizes animal species, but due to time constraints and this being out first hack, we could not add it to the app although the code is ready. The user will be able to take a picture with their camera in order to scan the animal and get an idea on what species it may be. One of the biggest challenges we had, besides using android studio, was trying to get google maps to feel snappy and responsive while in our app. Because google maps is such a huge file, it takes a lot just to build it let alone have it run at a decent rate in our app. We want the user experience to be as easy and clean as possible and a slow map would have definitely hindered that experience so it was essential we figured it out. Coming into this hackathon, both of us had any prior experience in UI or design, so to come in and make an app in less than 24 hours was a pretty big accomplishment. We learned a lot throughout the competition like the importance of and front end design, but most importantly It helped us realize how much you can accomplish with an idea, 24 hours, some laptops, access to the best API's and services, and a whole lot of caffeine. Now that we have the framework done, the next steps will be to train the model for better animal classification and add in more interactive features for the user to fell more involved. Optimizing is always important but I think a focus on user experience is one of the most important improvements to be made. We also hope to implement a dichotomous key for an even more accurate model along with also training the model to be more efficient and accurate.

Built With

• firebase
• flutter
• google-maps
• python