PointAction uses predicted dynamic 3D pointmaps from fine-tuned video models as an embodiment-agnostic action representation to map video predictions to executable robot actions.
Efficient4d: Fast dynamic 3d object generation from a single- view video.Int
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
PointAction: 3D Points as Universal Action Representations for Robot Control
PointAction uses predicted dynamic 3D pointmaps from fine-tuned video models as an embodiment-agnostic action representation to map video predictions to executable robot actions.