Inspiration
For the past month, we had been working on integrating the 200 USD AI depth camera from OpenCV with Rapyd API's. This will helps in easy checkout without the long queues.
What it does
There are AI depth cameras mounted on the entry and inside the store. When the customer enters the store, he/she can scan the QR using the grocery app, after scanning the unique customer id from the Rapyd API is collected and added with the id number allocated by the camera. The AI camera works on the edge, which means it is completely offline and doesn't need any internet connection(more secure). and doesn't use facial recognition. It uses the state of the art person re-identification AI model with more than 25 fps in real-time. When the customer takes a product it calculates the distance between the customer's palm and the product and identify it is taken or not. After it is taken the product gets updated with the unique id(customer id from Rapyd+ the id allocated by the camera) and can be visualized in the app cart, where easy checkout can be done by Rapyd collect.
How we built it
Built it using Dart, python, and weeks of research and debug into the code
Challenges we ran into
integrating the Rapyd API with the app
Log in or sign up for Devpost to join the conversation.