I am currently working on an object detection application for Android (written in Java) that uses Google ML. The application essentially focuses on drawing boundingBoxes around detected objects within a live streamed preview, provided by my device camera (via CameraX). To construct the base of this application, I used this CameraX documentation to build my camera view, and then this Google ML documentation.
But back to my issue in question; Here is what my application looks like when it detects an object. Excuse the fact that it's labelled as a letter opener. My biggest concern lies with how this issue can be resolved. It seems to be missing its detect object by a noticeable amount.
Here's the DrawGraphic java file that I used to draw the boundingBox:
Any further information required to supplement this question will be provided upon request.