Global Warming is a serious issue around the globe and not many people are aware of it. Huge amount of cars contribute a lot to global warming by carbon emission. As a driver, you never know how much carbon your car is emitting and that information is not easily available as well. By using this app, people may get more aware of how much damage they are doing to the environment.
What it does
This app uses deep learning to recognise which car is in the picture and then provides the CO2 emission in gram/kilometer to user.
How I built it.
I built an iOS app using Polarr's API and used core-ml model for recognising cars and also used European Unions research data on CO2 emission by cars .
Challenges I ran into
Using core-ml in iOS. I've some prior iOS app development but never used core-ml and Metal before. Also this app was made using API's made by Polarr only for this hackathon, so it was kind of challenging initially to use the API.
Accomplishments that I'm proud of
Learning to use core-ml on iOS and learning about many open source ML models available for future hacking.
What I learned
Other than working on my own project I actually helped 3 other teams in their project. A lot of teams consisted of Freshmen and I being a graduate student was more experienced with software development so I helped them and mentored them. I feel like I learnt a lot about how to be a good mentor. Also, I learnt how people with different background can play different role and help in the project.
What's next for carbon emission awareness
Currently I used a pertained model which is not super accurate, but I found some research papers and some car datasets to train machine learning models on. So, I'm going to train my own model and try to achieve higher accuracy. I would like to do more research on global warming and how it is going to affect millions of life on earth.