I'm a graduate student of Shanghai university.
I have read your paper about the part of Edge Loss Reinforced. That's a fascinating idea.
But I can't find the code about Edge Loss in this git.
I want to know, how do you get the edge coordinate and compute it.Is there a simple method?
Hi,
This implementation of ERN is based of Caffe, so you need prepare the edge GT data firstly, and then add corresponding data layer and loss layer in the "training.prototxt".
Hi,Liu
Best wish to You!
Li