SnapTrack

Logo
User Interface for Food Recognition

Inspiration

Nutrition facts labels are just too hard to read. Moreover, rarely do people bother to record and keep track of the amount of nutrition intake from these labels. Hence, a software that does that for people came to our mind.

What it does

It can take in a picture of either a nutrition facts table of food or simply the food and provide users with the nutrition breakdown with up to 16 kinds of nutrients. The result will be an approximation based on a serving size of 200g if the user inputs a picture of food. In addition, it has a built-in nutrition tracker that helps users track their daily nutrition intake with a customized scale.

How we built it

The project was made in Python with Pygame for visualization. We implemented Google Cloud Vision API for image and text recognition; we also used USDA Food Composition Database for detailed nutrition breakdown.

Challenges we ran into

Making animations and dynamic display of data.

Accomplishments that we're proud of

The whole project!

What we learned

How to use API to do cool things and achieve cool ideas!

What's next for SnapTrack

The next step is most likely to build it into an actual app with back-end database support and optimally have it also run on a mobile device, which will be more convenient and user-friendly.