Inspiration
Security cameras are everywhere. How can we make use of them better, while bringing the community together and making it a safer place for everyone?
What it does
Our web app receives footage of potential incidents that are then classified into actual incidents, and the authorities are notified, or normal events.
How we built it
We have a back end server written in flask Python which serves the front end (React JS, HTML and CSS). There is a node application that detects incidents using TensorFlow and collects the data from the Meraki API. Flagged entries are stored on the server with the metadata stored on the database.
Challenges we ran into
We encountered a problem with collecting data from Meraki API rather than from the actual camera itself. Moreover, the lack of documentation for Meraki API made it more challenging to develop our ideas.
Accomplishments that we're proud of
We have never worked together as a team so we are proud that we managed to come up with this idea and the final product by working collaboratively.
What we learned
The front end team was using React JS for the first time, and one of our backend members was using Flask for the first time. As a whole team, we learned a more pragmatic approach to system design.
What's next for au.pair
Our potential additions include improvements to the machine learning detecting various types of incidents.
Log in or sign up for Devpost to join the conversation.