We wanted to use the APIs that were available from the sponsors at hackNY, and we came up with this idea of being able to mash a few APIs together.
What it does
Receives images of food in text messages and orders whatever food you took a picture of for you
How we built it
Node.js and express app with Twilio, Clarifai, and EatStreet APIs to handle the texts, image recognition, and food ordering, respectively. The web app is hosted on AWS.
Challenges we ran into
Clarifai's image recognition is really general and you typically don't get the expected image tags for different types of food.
Accomplishments that we're proud of
We're proud to have finished merging the front end and back end in time for submission, and for successfully calling the appropriate APIs.
What we learned
It was my first exposure to Twilio and Clarifai, and none of us have ever used the EatStreet API before, so we got to explore a lot of new APIs in a single project.
What's next for Ordr•In
We plan on polishing our item detection algorithm in order to be more specific as to the type of food that is being reviewed. For example, fried chicken is both under chicken AND food, so it'd be better if we can pinpoint exactly what type of food it is that the user is looking for. We may also include options for the user to select how vague or specific they planned their picture to be, so they may choose from a wide range of options.