We wanted to build something that's related to ML/AI. Then with the theme, we got the idea to make something that connects two different systems. If we have a pet in the house, it's very hard to give time to look in CAMERAS all the time when you are away from home. That's the point, we decided to discover something that can identify the activity and sends the notification to the phone number.
It keeps track of pets at gate and notifies the client if they croses it.
We build the ML model using TensorFlow JS and implemented the notification (message) feature using Twilio.
Initially, we are finding it hard to implement the idea that we were thinking obviously with the time constraint. It was hard to find videos to test the model and build the connection between the flag generated with activity and the notification feature.
As soon as we get started gaining the confidence to improve and finish it all up on time.
TensorFlow JS, React JS, Node JS, Twilio. Most importantly this was our first ML/AL project. Lot of learnings!!
We'll try to add some extra features in the future.