The annotation of the image

[陈皓]

Dear Organizer,

Is there any ROI annotation for the pathological image?

Best,

Hao

[Eduardo Conde-Sousa]

Hi

No there isn't.

This is a binary problem without any annotations beside the label

All the best

Eduardo

--
You received this message because you are subscribed to the Google Groups "HEROHE Grand-Challenge" group.
To unsubscribe from this group and stop receiving emails from it, send an email to herohe-grand-chal...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/herohe-grand-challenge/6104a144-eeb6-4e19-bf8c-e82268dac7dd%40googlegroups.com.

[rob.k...@limebv.nl]

Dear Organizers,

I am not sure if an accurate binary classification is yet technologically possible in the way you intended, on a Giga-pixel size image. 

Are you aware of a similar competition (worth about 100 000 euro), https://tianchi.aliyun.com/competition/entrance/231757/information?lang=en-us, where the best result had only 0.30 mAP score?

Best,

Rob

Op vrijdag 29 november 2019 16:32:05 UTC+1 schreef econdesousa:

[Vincenzo Della Mea]

I am not among the organizers, however something "not yet technologically feasible" is perhaps the right topic for a challenge. However, there are at least a couple of papers where weak labeling of digital slides is successfully used.