# ScanQA

From the catalogue. Listed by one researcher from one source page.

- id: `scanqa`
- kind: language and reasoning annotation
- robot or device: ScanNet RGB-D indoor scans (class: sensor only)
- how collected: crowd annotation
- size as stated: over 40K question-answer pairs from 800 indoor scenes
- hours: not stated | hours only claimed: none | episodes: not stated
- year: 2021
- organisation: ATR-DBI (GitHub org)
- licence: CC BY-NC-SA 3.0 (class: non commercial)
- access: open (class: open)
- link (dataset): https://github.com/ATR-DBI/ScanQA
- note: Non-commercial licence; needs ScanNet.

Confirm the licence at the link before relying on it.