OCID-VLG

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

• id: ocid-vlg
• kind: grasping
• robot or device: OCID scenes with referring text, masks and grasps (class: sensor only)
• how collected: annotation on OCID
• size as stated: 89,639 image-text-mask-grasp tuples from 1,763 scenes
• hours: not stated | hours only claimed: none | episodes: not stated
• year: 2023
• organisation: University of Groningen
• licence: not stated (class: not stated)
• access: open (class: open)
• link (dataset): https://github.com/gtziafas/OCID-VLG
• note: Vision-language-grasping in clutter.

Confirm the licence at the link before relying on it.