Inspiration

We were inspired to make this web application in order to make the weekly task of buying groceries easier and hence bring about some positive change upon the lives of others

What it does

Checkout.io was made to make the grocery shopping experience hassle free. Once a customer has finished shopping, they simply place their shopping cart in front of a camera connected to Checkout.io. The application will then return the name and price of all the items in the customers shopping cart.

How we built it

The front end was built using JavaScript while the back end was built using Django Framework. On the client side, jQuery was used to send AJAX calls to the server. On the server side, we used Tensor Flow to create a machine learning model for object detection. In addition, we also incorporated Wegmans API which allowed us to return the price of items in their store.

Challenges we ran into

Integrating all the different parts of our system was challenging. One such instance included us having to figure out how to send data from the front end to the back end so that it could be processed by the Tensor Flow Models.

Accomplishments that we're proud of

One of the accomplishments that we are very proud of is our ML model. It was able to differentiate with a high level of certainty between different items such as coke cans and water bottles. We tested with these items as they can be found in almost every grocery store.

What we learned

We learned a lot about system integration and about data formats that allow data to flow from client to server, such as JSON and XML.

What's next for Checkout.io

The next step for Checkout.io would be to better perfect our model for detecting objects. Ideally, we would like to have our application require the shopper to make no adjustments in the positioning of items in their cart, hence requiring our application to pick up on items that are overlapping.

Built With

Share this project:

Updates