Questation about the multi-class for the "High Quality Segmentation for Ultra High-resolution Images"

Open Ianresearch opened this issue 3 years ago • 4 comments

Hello! Have you tried to refine the mask which is more than one class at same time, such as 3 classes or 4 classes? And how about the result? Thank you very much.

May 05 '22 03:05 Ianresearch

The model is only in class-agnostic style now. Thanks.

May 05 '22 03:05 tcShen

Thank you for reply. And I have another queatation.'The perturbed masks are randomly perturbed on ground truth with a random IoU threshold between 0.8 and 1.0' -- can you explain the perturb method more detail,and open the code.Look forward your reply.

May 12 '22 02:05 Ianresearch

https://github.com/dvlab-research/Entity/blob/main/High-Quality-Segmention/util/boundary_modification.py The perturbation code is in the link above. You can take several minutes to read it. Thanks.

May 12 '22 07:05 tcShen

OK. Thank you very much!

May 12 '22 08:05 Ianresearch