Hi guys, I am using camera and firebase_ml_vision to implement text recognize, but this current api apply the whole ImageStream to firebase. In my app, I only want to feed part of ImageStream Example. How do I do this?
Options I have tried:
Option 1: Crop CameraImage data in startImageStream. Don't know how to do it.
Option 2: FirebaseVisionImageMetadata only have size property, don't have position.
Option 3: Remove TextElement which not in the detect area and manually rebuild recognized text after detected text. Too complicated and can have a performance issue.