Hey! This is my implemention of an intelligent camera that can detect objects and automatically zoom to the selected object!
It was built as a qualification task challenge for https://ccextractor.org/
This application uses google_mlkit_object_detection with a custom trained model to detect objects. When it detects an object, it draws a rectangle around it (bounding box) using CustomPainter.
When you touch the rectangle, it will automatically zoom to fit the object!
The application also allows you to reset zoom, take a picture (saves to your gallery) and pause/play the camera preview!
You can find the first pre-release here!