Swamphacks X Hackathon 1/25/2024 - 1/26/2024
Project: Scan To Recipe Contributors: Krish Shah, Andrew Vu, Sivan Pushpagiri, Jovon Alexis
Food waste is one of the biggest epidemics in the world, yet the cause of this is even more tragic. Food ends up in the trash not because of malice, but due to lack of motivation and knowledge in how to use them. This is what Scan To Recipe aims to fix.
Scan To Go is a website that scans random ingredients a user may have. When the user is done, the website will generate different recipes based on ingredients.
We first created spent time creating the website, adding the different pages we wanted and how we wanted it took. We then had to access the user's camera, so it could snap a picture and analyze the picture for any food. After this is finished, we collect the detected items and recipes are generated. **We made us of ChatGPT and Perplexity during this project.
This biggest challenge we had was finding a way to scan/detect objects. We tried using combinations of AI, pre-trained AR models, and different APIs in order to do it, but it didn't turn out as we hoped. We eventually ended up using manually inputting ingredients for the model to look for, and it ended up working to our liking.
Getting the scan to actually work. It was the hardest part about this project, and definitely took the longest. While it doesn't do all that we imagined, it get its job done. Considering this is all of our team's first hackathon, we're very proud.
How to use teamwork to our advantage, specifically divide and conquer. We also learned that research is very important to get anything done, as there were times where we were lost. We also realized that it's better to plan things out, rather than just dive straight in when an idea comes.
**This project was created with the help of ChatGPT and Perplexity