So ideally the ml network would solve the problem end to end but these authors seem to be using a network for only one step of their otherwise classic image processing pipeline
A lot of applied work in vision and audio glues together different existing modules, instead of training the whole thing end-to-end.
In an ideal world, things are ideal. But the world isn't a grad student's wet dream.
But there are not as many influential papers in the world as there are people who respond: "Why didn't they train the whole thing end-to-end?"
So ideally the ml network would solve the problem end to end but these authors seem to be using a network for only one step of their otherwise classic image processing pipeline