Have you attempted to perform weakly-supervised semantic edge extraction?

[Yancccccc]

I would like to combine your article with the implementation of semantic edge extraction using CASENet, but I'm not sure how to generate pseudo labels.

[voidrank]

Hi @Yancccccc

The only concern is that, in DiscoBox or MAL, the training data is fed in a way that only one object is in the center of the image. If your semantic segmentation data meets the requirements, it should be easy

Shiyi