Inspiration
We love pasta nights, but it's frustrating when we're missing key ingredients for our favorite recipes. This always leads to compromised pasta experiences. On top of that, we hate wasting food. We envisioned an app that would help us cook amazing pasta dishes with whatever we have on hand so no good food goes to waste.
What it does
PastaMe allows users to take photos of their available ingredients. It analyzes the images and suggests tailored pasta recipes using only those ingredients. The app provides easy-to-follow instructions for delicious pasta creations.
How we built it
We built PastaMe as a mobile web app using React. Users upload photos which are processed by the Google Vision API to identify ingredients. This ingredient list is sent to the Edamam recipe API, filtered for pasta dishes. The app displays recipe results that perfectly match the user's on-hand ingredients. We chose to build it as a web app rather than a native mobile app to make it more accessible across devices and platforms. For testing we opened the app on our desktop browsers, but set the responsive dimensions to "iphone 12 pro".
Challenges we ran into
The main challenge we faced was getting detailed and accurate ingredient identification from the Google Vision API. We struggled to get the full information we needed rather than a more generic analysis. However, after much trial and error, we found techniques to improve the vision detection and get it working well.
Accomplishments that we're proud of
We are really proud that we overcame the hurdle of getting usable ingredient data from the vision API. Figuring out how to optimize the image processing was a major accomplishment.
What we learned
This project taught us a lot about both computer vision and working with external APIs. We gained valuable experience in debugging APIs and learning how to adjust our inputs and processes to get better results.
What's next for PastaMe
The next step for PastaMe is expansion and refinement. We want to add more cuisines beyond pasta, improve the ingredient detection accuracy, and create a smoother user experience. More customization options and recipe recommendations are also on the horizon.
Built With
- api
- axios
- bootstrap
- css3
- edamam-nutrition
- express.js
- figma
- firebase
- github
- google-cloud
- google-vision-api
- html5
- javascript
- json
- jsx
- node.js
- python
- react
- react-router
- reactstrap

Log in or sign up for Devpost to join the conversation.