FCBV-Net achieves superior category-level generalization for bimanual garment smoothing by conditioning value prediction on static pre-trained dense geometric features from point clouds, showing only 11.5% efficiency drop on unseen garments versus much larger drops in baselines.
Towards building ai-cps with nvidia isaac sim: An in dustrial benchmark and case study for robotics manipulation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
FCBV-Net: Category-Level Robotic Garment Smoothing via Feature-Conditioned Bimanual Value Prediction
FCBV-Net achieves superior category-level generalization for bimanual garment smoothing by conditioning value prediction on static pre-trained dense geometric features from point clouds, showing only 11.5% efficiency drop on unseen garments versus much larger drops in baselines.