ATLAS-120k

ATLAS-120k Surgical Anatomy Recognition with Context Learning using Foundation Representations ATLAS-120k is a large-scale clip-level semantic segmentation dataset for surgical anatomy recognition in minimally invasive surgery (MIS). It comprises over 120,000 annotated frames from 100 surgical videos spanning 14 procedures and 42 anatomical classes, covering both laparoscopic and robot-assisted surgery. Restricted post-release review period — until 31 July 2026.… See the full description on the dataset page:

Source: Hugging Face Hub (TimJaspersTue/ATLAS-120k). Metadata imported from the dataset’s Hub tags.