enter image description here

I am trying to implement this paper


Here they have mentioned text local (edge) and global regions as supervisory information. Side output is generated and then this supervisory information is introduced as shown in the image. I cannot understand how supervisory information is introduced. How to train through side output and how to make the network recognize the features. I am trying a keras implementation. Any relevant article or similar projects will be of great help.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.