Poor Dice score in polyp segmentation #38

Open

Open

Poor Dice score in polyp segmentation#38

mostafij-rahman

UNeXt produces surprisingly poor Dice scores (around 80% while SOTA is around 93%) for polyp segmentation using the Kvasir-Seg dataset. Is there any way to improve?

Metadata

Assignees

No one assigned

Labels

No labels

No labels

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests