How to offline visual recognition (features like MLCore, but cross platform)

Hi

I'm looking for ways to implement object recognition into a cross platform Unity app.
I have been experimenting with Apple CoreML before and it is sufficient for my goal, except that it only works on iOS.

Any tips on a cross platform visual recognition SDK?

Firebase ML-kit looks very promising but lacks Unity integration.
https://firebase.google.com/docs/ml-kit/