InsTags

LandingPage

Inspiration

What it does

Generates relevant, realtime hashtags from just an image using Google Cloud's Vision API's image recognition technology for use on any social media platform.

How we built it

The general process is 1) send the selected image to multiple AI image-recognition service providers (Google Vision and Imagga), 2) the backend server receives the data from the providers, aggregates the results based on our algorithm that filters the "words" by context, 3) finally, all the selected "words" are transferred over to the front-end and is displayed accordingly on the app.

Challenges we ran into

Angular deployment to server was problematic due to some compression issue. Also, most of the social APIs that we were going to use (i.e. Facebook, Twitter, Instagram) were either limited, or were tied to deprecated services that hindered us from getting richer and more accurate data for the users.

Accomplishments that we're proud of

We were able to come up with the idea and FOCUS on what mattered (core function of the app) and what did not (using other APIs). This saved us time and allowed us to streamline our process of creating the app all the way from the front to the back. Also, linking up all the APIs that were a first-time to us really was an accomplishment for us.

What we learned

We learned that the Google Vision API is really powerful. Having said that it wasn't enough to provide more rich data that had context, thus, looking into Imagga as well. We also learned that however many times we plan things, time is always against us, and we need to efficiently manage the time to produce a viable working product.