# DL3DV-10K

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

- id: `dl3dv-10k`
- kind: world model and video
- robot or device: handheld scene capture video (class: sensor only)
- how collected: human filmed real scenes
- size as stated: 51.2M frames, 10,510 videos, 65 location types
- hours: not stated | hours only claimed: none | episodes: not stated
- year: 2023
- organisation: DL3DV team
- licence: not stated (class: not stated)
- access: gated (class: gated)
- link (dataset): https://huggingface.co/datasets/DL3DV/DL3DV-ALL-960P
- Hugging Face repository: `DL3DV/DL3DV-ALL-960P`, 101,078 downloads in 30 days, https://huggingface.co/datasets/DL3DV/DL3DV-ALL-960P
- note: Widely used for camera-controlled world model training.

Confirm the licence at the link before relying on it.