foodRecc

Inspiration

We love food and wanted to try harnessing the power of Google Cloud Vision to discover new restaurants

What it does

foodRecc takes photos from a user's Twitter feed and recommends restaurants based on photos of food

How we built it

We built foodRecc using Google Cloud Vision, Twitter, and Zomato APIs

Challenges we ran into

We initially wanted to get photos from Instagram accounts, however Instagram recently changed their cryptography and made it more difficult to scrape photos

Accomplishments that we're proud of

Completing this project, using 3 APIs simultaneously together in a smooth manner in the span of 24 hours.

What we learned

Setting up a good development environment and installing Python dependencies can be difficult at times. Instagram is not very developer friendly and actively makes it difficult to scrape images from their platform.

What's next for foodRecc

The world

In all seriousness, we would like to get Instagram integration and improve the recommendation system to be less naive

Built With

Submitted to

BCHacks 2020
- Winner 3rd Place

Created by

I worked on the backend working with the Google Cloud Vision, Twitter, and Zomato APIs. This was my first Hackathon and it was really great experience.

Darryl Lam
I worked the front end. All the ui was made by me. It was my first time doing front end work and setting up dependencies and getting all the python versions and things to work together is a very big pain. The main thing I struggled with was visualizing how the ui would look when I was typing it up. Getting the back end to work with the front end was a challenge but we overcame it

Nachiketa-k Kumar
I worked on the Instagram/Facebook API which turned out to be a bust due to the Instagram API not being as developer friendly as we may have wished. Due to this, we substituted for other API's that were able to fill our needs instead. I also worked on Human-Computer interaction in the back-end of the project as well as overseeing the project by managing the group.

Kyle Cantlon
Computer Science Connoisseur
I worked on the implementation of live stream video capture from mobile phone and process shots each time the webpage is refreshed. Then, those shots could have been analysed by the google's vision API. Unfortunately, the API didn't work properly with the livestream. I also helped for the frontend development.This was my first time working with python. It was a wonderful experience working with the team.

Narshvin Doorganah