New recipes are really fun, and sometimes you don’t get many of them. We wanted to see if we could build something to find new recipes for us. However, there was one more problem, we always had to go to the store to buy something or the other. Now, we can find recipes without having to get much (or none) from the store!
What it does
Our solution is the use the Google Vision Api and one of the best food api’s on this planet, spoonacular, to make a web-app that allows you to input a photo of some ingredients and it will output recommended recipes. In Google Vision Api we had to use labels, ocr scanning for canned products, as well as object detection. For spoonacular we had to make many calls to get recipes and info. These backend functions come and work in sync with the frontend to deliver many recipes quickly. No more flipping through cookbooks and online articles for something you can make, CyberChef is here to show you what to cook and bake!
How we built it
Challenges we ran into
- CSS - we spent five hours fixing and changing our User Interface.
- Google Vision API - some results Google Vision gave us were inaccurate and we needed to fix it
- Spoonacular API - It kept giving us errors as we were coding it
- Teamwork - our team got into a few disagreements about what the site should do
- Communication - it was hard to get our whole team online at once and for them to join the Zoom meetings and Discord Voice Chats
- Replit - replit, our hosting site, had some package installing errors that we had to deal with, we decided to use digital oceans in the end
Accomplishments that we're proud of
- We learned about many food APIs
What we learned
- We learned many APIs related to food.
What's next for Recipe Finder
- We will add more features such as Restaurant search
Log in or sign up for Devpost to join the conversation.