Object Detection ML Package Documentation Question

I’m a bit confused by this section of the documentation in Red on this page:

What is the “above image” and the “Mask” that are being referred to in this single line of documentation? :thinking:

Thanks!