any details regarding the experiments on the kvasir dataset?

tianrun-chen / SAM-Adapter-PyTorch

Adapting Meta AI's Segment Anything to Downstream Tasks with Adapters and Prompts

MIT License

1k stars 89 forks source link

any details regarding the experiments on the kvasir dataset? #32

Closed uobinxiao closed 1 year ago

uobinxiao commented 1 year ago

Could you elaborate a little bit the evaluation protocol on the kavisr dataset? The data splition and others.

tianrun-chen commented 1 year ago

Hello! As we have mentioned in the main manuscript, we use the same protocol (including the test-train split) from Medico multimedia task at mediaeval 2020: Automatic polyp segmentation.

uobinxiao commented 1 year ago

thanks for your reply. I want to list your result as benchmark, but the test set of kavisr dataset doesn't contain ground truth. Did you send your results to the organizer to get the scores reported in your paper?