Vuzii

Inspiration

• We drew inspiration from similar image sharing sites like Instagram and Flickr.
• Since our challenge was based on Vuzii, who also implicitly drew inspiration from them.

What it does

• Images are loaded from an API provider, for example Vuzii or Instagram
• The Google Cloud Vision Neural Network is used to tag items in images
• Relationships are determined between images using our Linkify(TM) technology
• The resulting data is displayed in a network visualisation, organised based on relationships

Challenges we ran into

• We were instructed to use the Instagram API as a source of data, but we could not gain access to public image feeds without waiting for our developer application to be processed.
• Instead, we had to generate mock data to serve in place of real Instagram data.

Accomplishments that we're proud of

• It works!
• Managed to extract meaningful data from images using Google Cloud Vision API.
• Pizza!

What we learned

• We learned how to use Google's Cloud Vision API for identifying objects on an image.
• We learned how to visualise a node-based data structure using the vis.js JavaScript library.
• We learned how to work on a full-stack Java application in a team.

What's next for Vuzii

• It would be nice to actually get it working with the Instagram API instead of working with fake data.
• We could pull images from more image-sharing services like Flickr and Pinterest.
• We could tag regions of the image with links for online shopping using the Cloud Vision API, and perhaps Amazon and/or Google Shopping.

Built With

• css3
• google-cloud-vision-api
• gradle
• html5
• intellij-idea
• java
• javascript
• jquery
• lombok
• node.js
• vis.js