Thank you for sharing this great dataset!
I downloaded the dataset from Version COCO: https://zenodo.org/records/10390295 and noticed that there is some duplication in annotation IDs in arcade/stenosis/train/annotations/train.json. You can find it by searching for "id": 279.
Shouldn't annotation IDs be unique?
Thank you for sharing this great dataset! I downloaded the dataset from Version COCO: https://zenodo.org/records/10390295 and noticed that there is some duplication in annotation IDs in arcade/stenosis/train/annotations/train.json. You can find it by searching for "id": 279. Shouldn't annotation IDs be unique?