search

Bedrettin-Cetinkaya / RankED

MIT License
19 stars 3 forks source link

About Section 3.1. Ranking Positive Edge Pixels over Negatives in paper #1

Closed Ozone-oo closed 7 months ago

Ozone-oo commented 7 months ago

I've been working on replicating the findings from your paper, "RankED: Addressing Imbalance and Uncertainty in Edge Detection Using Ranking-based Losses", specifically the method described in "Section 3.1. Ranking Positive Edge Pixels over Negatives". I've hit some roadblocks and would greatly appreciate if you could release the code pertaining to this section.

Thank you for considering my request. I'm open to any further dialogue and ready to adhere to any terms you might set for sharing the code.

Bedrettin-Cetinkaya commented 7 months ago

Thank you for your interest. I will share code and pretrained models asap.

Bedrettin-Cetinkaya commented 7 months ago

I added neccessary files for training & inference. Evaluation code and our pretrained model will be uploaded asap.