libero-memMIT
LIBERO-Mem: Long-Horizon Object-Centric Kitchen Manipulation
A dataset of long-horizon object-centric kitchen manipulation tasks with synchronized RGB-D observations, segmentation maps, proprioception, and control actions from dual camera viewpoints.
Downloads191
Likes3
Why This Matters for Physical AI
LIBERO-Mem provides long-horizon manipulation demonstrations with object-centric annotations and dual-view observations, enabling research in imitation learning and reinforcement learning for complex kitchen manipulation tasks.
Technical Profile
- Modalities
- rgbdepthproprioceptionsegmentation
- Robot Embodiments
- Franka Panda
- Action Space
- joint_positions
- Environment
- kitchensimulation
- Task Types
- manipulationpick_and_placeobject_rearrangement
- Data Format
- HDF5
- Annotation Types
- language_instructionsbounding_boxessegmentationreward_labels
- License
- MIT
Community Signals
Top 50% by downloads
Access
Need custom rgb data?
Claru builds purpose-built datasets for kitchen applications with dense human annotations and quality assurance.
Request a Sample Pack