I always thought it was cool to be able to evaluate or price everyday objects that you both own or don't, just to know how much you could sell / buy them for. This app is our solution to make this as interactive and easy as possible.
What it does
With augmented reality we can scan objects using image recognition around your room. The app then displays the name of the object, the average retail price on Walmart, and the highest big price on eBay. This app is designed to show you the value in objects you never considered.
There is language support so that after identification of an object, it can be pronounced and shown to you in any language.
How we built it
The core of our project, is our Python REST API. We used Unity3D to process capture the image, where it was then exported to our python server. The server acts as router that directs the image to the Cloudsight API, which processes and returns a full phrase associated with the scanned image. Using tailored word processing, the data is then sent through both Walmart's and eBay's API to acquire a retail value and highest bid value of the scanned object. In the background, our images are stored using Parse, and our web server is managed through Microsoft Azure.
Challenges we ran into
To link the c# Unity3D client with the python server APIs, as well as learning to configure the meta goggles themselves were all challenges in their own right. After hours of scratching our heads, we can happily say we solved almost all of these challenges.
Accomplishments that we're proud of
Working together to making something we all were passionate about. Also overcoming the long and hard to find bugs that almost prevented us from finishing this project.
What we learned
We learned more about almost every API mentioned earlier. Most were not common to us, however after this hackathon we can see we are one step closer to being able to use some of these APIs to much greater extents.
What's next for e-Valuator
More platforms, since the REST API is reusable.