Cloth-splattersMIT
dexgarmentlab-folding-lifting-meshes
Simulated garment-manipulation demonstrations for training cloth state-estimation and dynamics models on variable-vertex meshes, recorded with DexGarmentLab assets.
Downloads0
Episodes894
Why This Matters for Physical AI
This dataset enables training of cloth state-estimation and dynamics models on deformable objects with variable mesh topologies, which is critical for advancing manipulation of soft materials in robotics.
Technical Profile
- Modalities
- point_cloudproprioception
- Robot Embodiments
- gripper
- Environment
- simulation
- Task Types
- cloth_manipulationfoldinglifting
- Episodes
- 894
- Data Format
- HDF5
- Annotation Types
- mesh_topologyvertex_positionsgrasp_labels
- License
- MIT
Access
Need custom point_cloud data?
Claru builds purpose-built datasets for simulation applications with dense human annotations and quality assurance.
Request a Sample Pack