Inspiration

Everyone struggles to find that perfect caption to a little spice to their Instagram posts. CapGen makes it easier by generating relevant captions from public quotes datasets using Google's Vision API.

What it does

CapGen uses the Google Vision API to analyze image data and produce certain keywords that are then used to produce random quotes for Instagram captions.

How I built it

We built our app using Android Studio and Java. To mock up the wireframe for our app, we used Figma. For the image recognition, we used Google’s Vision API, and for quote generation, we sourced quotes, lyrics, and facts from several public datasets such as Wikiquote.

Challenges I ran into

We found too large of a dataset for Github to handle. We also had trouble integrating “Import from Photo Gallery” functionality. However, we overcame those challenges.

Accomplishments that I'm proud of

Adam: Working with a Google API and with Android Studio, even though I've used neither before.

Aaron: Learning how to use Figma and working on a project that uses large data sets effectively.

Cole: Writing Java scripts that access public quote APIs and finding and creating quality data sets.

What I learned

Through this experience, we all learned about the power of visual recognition such as Google’s Vision API. We also have a better understanding of what it takes to integrate functionality such as that into an app even as simple as ours. We also maintained a level of camaraderie that helped us to develop solutions for many of our technical and semantic issues.

What's next for CapGen

We plan on developing CapGen further based on the interest shown in it during SwampHacks. We have plans to create a better user interface and to allow for direct uploading of captions to Instagram.

Built With

Share this project:

Updates