People love taking pictures of their food. But they don’t often think about the carbon impact of their food choices, or of the overall world food system. We wanted to make an app that would help people make eco-conscious choices when it comes to the food they eat, and also to engage them in thinking critically about where their food comes from.
We decided to create a machine learning based app that would analyze user uploaded photos. The key idea is to make it a fun experience to use - and we incorporated humor and storytelling to that end.
What it does
Users can upload a photo of their meal, and the app will give them thought-provoking information about the carbon-impact of ingredients of their meal.
How we built it
We trained Google's Cloud Vision to recognize various dishes, and we built the website using Flask.
Challenges we ran into
There were certain categories of data that were harder for the model to distinguish between. Our biggest stumbling block was having trouble connecting the vision data to our webpage, we had trouble getting OAuth to authenticate properly.
Accomplishments that we're proud of
We were able to successfully train a model and use machine learning for an environmental-oriented app.
What we learned
We learned how to train a model with google cloud, and more about various vectors for the carbon impact of food.
What's next for our app
Integration of more categories of food.