# DexArt

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

- id: `dexart`
- kind: teleop
- robot or device: multi-finger robot hand with articulated objects in simulation (class: dexterous hand)
- how collected: no demonstrations: reinforcement learning (PPO with PointNet) only
- size as stated: no demonstration dataset
- hours: not stated | hours only claimed: none | episodes: not stated
- year: 2023
- organisation: UC San Diego
- licence: not stated (class: not stated)
- access: open (class: open)
- link (dataset): https://www.chenbao.tech/dexart/
- note: RL benchmark. No human or teleoperated data.

Confirm the licence at the link before relying on it.