Hi. I’m hoping that someone can point me in the correct direction please. There is a Smart Detection "Cross Line’’ capability in UniFi Protect, where a smart (person/vehicle/animal) detection is made if crossing a line created in the protect camera config.
I’d like to get this Cross Line detection in Home Assistant, separate from other smart/person/vehicle detection (i.e. to know that the vehicle is inside the driveway, rather than outside). Is this possible please?
Note: I do have smart detections working in Home Assistant already, I’m hoping there is a way to separate Cross Line detections from general smart detections. I have searched around for info on this, but can’t seem to find …
If anyone else was confused about the state of the latest UniFi versions and the compatibility with the HA integration, this GitHub PR should help clear things up.
TL;DR:
UniFi OS 4.x : Incompatible
UniFi Protect 4.x: Compatible, with caveat
This is also worth a read as it provides some background.
I read about the solcast integration maintainer’s quit yesterday, and now this… shame, two integrations I use a lot. I’ve had “interesting” interactions with both, so I’m not surprised to see the behaviours.
In my experience over more years than I care to think about, individual contributors often think they are irreplaceable, but rarely are, once the initial change shock passes.
Mine was also working fine with UnifiOS4.0.6 until I updated Home Assistant to 2024.7.1 and now all my object, person, and vehicle detections are unavailable. It’s broken so many of my automations. I used all my cameras as sensors for lighting. The whole integration is useless now. I think I’ll have to roll back.
Hoping someone can help me out here. I want to use the Unifi cameras ability to AI detect smoke detectors to set up a smoke detector in home assistant.
Background: I have a bunch of G5 Flex cameras outside. The only camera inside is one G4 Instant in my young child’s room.
In the Unifi interface, both the G5 Flex and G4 Instant can detect smoke/CO alarms:
