Skip to content

Dataset · Aeoo

Colonoscopy polyp images and masks (CVC-ClinicDB)

Aeoo/CVC-ClinicDB

612 medical frames from 29 colonoscopy recordings with binary masks for polyp segmentation.

What's inside

A specialized medical dataset for binary image segmentation (polyp vs. background). Contains 612 PNG frames of 384x288 pixels extracted from 29 colonoscopy video sequences, along with binary mask annotations. Data is already split into training, validation, and test sets.

What you can do with it

  • Training and fine-tuning computer vision models (U-Net, SegFormer, SAM) for medical diagnostics
  • Benchmarking automated neoplasm detection algorithms on endoscopic images
Tasks
Image segmentation
Size
up to 1K rows
Downloads this month
23.9K
Likes
2

How to load it

from datasets import load_dataset

ds = load_dataset("Aeoo/CVC-ClinicDB")