Inspiration

Food is an important part of our daily lives. With the assistance of machine learning, the ability to identify unknown foods through image recognition opens new doors for culinary enthusiasts who desire to seek out new local restaurants and recipes, expanding their world of food, one picture at a time.

What it does

The user is allowed to take a picture of a complete dish or specify a list of basics ingredients. Alternatively, a user may also input a keyword. Utilizing machine learning using Convolutional Neural Network, the program will first identify the picture sent. For complete dishes, the program will display local restaurants that serve the item, or recipes to make the dish. On the other hand for ingredients, the program will analyze what ingredients you have, and return a recipe which caters best to what is available in your kitchen.

How we built it

As a team, we decided to create an iOS and cross platform Android mainly via react-native. API calls were made using the google-places API for restaurant locations, and the Spoonacular API for recipes based on either dishes or given ingredients.

Challenges we ran into

Early on, during the project, we ran into issues connecting the mobile app to our own server's API.

Accomplishments that We're proud of

An accomplishment that we can be proud of is that we managed to finish the project for the Western Hackathon within the designated time frame.

What we learned

During the endeavor, we learned to manage time and follow group protocol in order to have every individual member on the same page. We also learned the importance of understanding each other's strengths and weakness because a team is only as strong as its weakest link. Knowing each other's weaknesses mitigates this fact as teammates can cover for each other's weaknesses and work together to further enhance the group's overall strength.

What's next for ChowDown!

Integrating additional data sets into machine learning for the image recognition algorithm in order to increase its success rate of identifying certain types of foods.

Share this project:
×

Updates