This release of the app preforms all required tasks at a very basic level. Some features, like object detection and boundary boxes, have not been implemented yet due to API version conflicts. So currently the app uses image classification.
Download the APK and luanch.