oxe-auge2025cc-by-4.0
language_table_train_10000_15000_augmented
An augmented subset of the Language Table dataset containing 5,000 episodes with multi-robot cross-embodiment visual and proprioceptive observations paired with natural language instructions. The dataset provides augmented images and trajectories for 8 different robot embodiments (Google Robot, Jaco, Kinova3, KUKA IIWA, Franka Panda, Sawyer, UR5e) to enable scaling of cross-embodiment policy learning.
Downloads2K
Episodes5000
Hours2.2136
Technical Profile
- Modalities
- rgbproprioceptionlanguage
- Robot Embodiments
- Google RobotJacoKinova3KUKA IIWAFranka PandaSawyerUR5e
- Environment
- lab
- Task Types
- manipulation
- Episodes
- 5000
- Total Hours
- 2.2136
- Data Format
- parquet
- License
- cc-by-4.0
Community Signals
Top 50% by downloads
Access
Need custom rgb data?
Claru builds purpose-built datasets for lab applications with dense human annotations and quality assurance.
Request a Sample Pack