[firebase_ml_vision] Face detection In the Cloud, Custom Models and What is the Image Labeler?

[bansalarnav]

Hi,
So i had some common questions that i guess many others have too. Would appreciate it if someone could come forward and answer all of these questions -

1. Currently if we follow the given example it does on-device Face Detection. So it bundles the model with the app which increases the app size to quite some extent. Is their any way to perform face detection on the cloud so it doesnt bundle model with the app? I heard that firebase has a cloud api for face detection.

2. If i have uploaded a custom .tflite model to firebase is their any way through which i can perform inference with in in my Flutter app? Currently i can only do this by adding the model to my app assets and using the other tflite plugin. I want to be able to download the latest version from firebase and perform inference on it.

3. What exactly does the Image Labeler do? Does it just determine the confidence level of a human in a picture? Or can we use any model that we uploaded to the firebase console?

Would appreciate it if someone could explain the above 3 questions and if possible give code examples too.
Thanks