This is a Home Assistant custom component for Lobe. Lobe is an AI tool that can classify images. This component lets you easily use an exported model along with another server to classify a camera entity's feed with it.
Use HACS for the integration. You'll also need a seperate server. Steps to install on another server:
- Install the Lobe library.
- Install Flask.
- Export a Tensorflow Lite model into a folder on the server.
- Copy over app.py and change the folder location.
- Run app.py.
- You'll probably want to make it run on start.
This is the configuration format:
image_processing: - platform: lobe entity_id: camera.front_door_livestream # Camera entity ID name: "Front Door Status" # Optional; Custom name server: "http://lobeserver.local:5623" # Server address scan_interval: 2 # Optional; How often to update
It will produce an entity something like this: