EndoVis

EndoVis is a family of challenge datasets for endoscopic computer vision across segmentation, detection, localization, reconstruction, and related tasks. It is a valuable reference for surgical perception, but access and statistics vary by sub-challenge.

• Challenge-specific endoscopic datasets
• Computer-vision labels such as segmentation, detection, depth, or localization
• Benchmark rules and evaluation protocols

EndoVis organizes public challenge tasks using endoscopic data released under sub-challenge-specific access rules.

• Surgical scene perception
• Tool segmentation
• Depth and reconstruction
• Computer-assisted intervention benchmarks

This is a third-party public/research dataset. We do not own or distribute this dataset. Access and usage are governed by the original publisher's terms.

Access and use are governed by each EndoVis sub-challenge's rules and publisher terms.

Custom healthcare data may be needed when teams require non-endoscopic workflows, synchronized robot actions, or care-environment demonstrations.