policy
CLIPort
UW / NVIDIA / Google · PyTorch
or hover any field below to flag it
Overview
Name
CLIPort
Author
UW / NVIDIA / Google
Framework
PyTorch
License
apache-2.0
Skill type
manipulation
Evidence level
verified
Task description
Language-conditioned imitation learning combining CLIP semantic understanding with TransporterNet spatial precision. Two-stream architecture processes RGB-D observations to predict pick-and-place actions via pixel locations and rotations. Handles 10+ tasks from language instructions.
Spaces
Action space
other · 6-dim · 1Hz
Observation space
- type: multimodal
- · rgbd_topdown
- · language_instruction
Links
HuggingFace repo
null
Paper (arXiv)
Compatible robots
0+2 mentioned but not in catalog yetNo robots list CLIPort as compatible yet. Know of one? Flag it above.
Compatible environments
1Datasets that reference this policy
0No datasets reference CLIPort yet.