Dataset · saadwazir
Medical segmentation dataset collection (ECCV 2026)
saadwazir/MedCAGD-Dataset-Collection
From 10 to 100 thousand medical images for training and evaluating segmentation models.
What's inside
A collection of public datasets for semantic segmentation of medical images, prepared for the MedCAGD (ECCV 2026) work. Includes from 10,000 to 100,000 images laid out in subfolders in their original or standardized structure. Accompanied by English metadata.
What you can do with it
- Training and comparing medical image semantic segmentation models
- Reproducing experiments and benchmarks from the MedCAGD paper
- Tasks
- Image segmentation
- Languages
- English
- Size
- 10K–100K rows
- Downloads this month
- 35.2K
- Likes
- 1
How to load it
from datasets import load_dataset
ds = load_dataset("saadwazir/MedCAGD-Dataset-Collection")