HUG trains a flow-matching model on a new 1M-frame egocentric human grasp dataset to generate retargetable grasps from single RGB-D images, beating baselines by 23-34% on a new 90-object benchmark.
AnyDexGrasp: General dexterous grasping for different hands with human-level learning efficiency
9 Pith papers cite this work. Polarity classification is still indexing.
9
Pith papers citing it
abstract
We introduce an efficient approach for learning dexterous grasping with minimal data, advancing robotic manipulation capabilities across different robotic hands. Unlike traditional methods that require millions of grasp labels for each robotic hand, our method achieves high performance with human-level learning efficiency: only hundreds of grasp attempts on 40 training objects. The approach separates the grasping process into two stages: first, a universal model maps scene geometry to intermediate contact-centric grasp representations, independent of specific robotic hands. Next, a unique grasp decision model is trained for each robotic hand through real-world trial and error, translating these representations into final grasp poses. Our results show a grasp success rate of 75-95\% across three different robotic hands in real-world cluttered environments with over 150 novel objects, improving to 80-98\% with increased training objects. This adaptable method demonstrates promising applications for humanoid robots, prosthetics, and other domains requiring robust, versatile robotic manipulation.
fields
cs.RO 9representative citing papers
A hardware-free dual-camera capture framework with ChArUco spatial unification and receding-horizon state alignment enables decoupled SE(3) manipulation and SE(2) base trajectories for diffusion policies, yielding 83.8% average success on four long-horizon household tasks.
BiDexGrasp supplies a 9.7-million-grasp bimanual dexterous dataset built via two-stage synthesis and a coordinated geometry-size-adaptive model that generates grasps for unseen objects.
LACE aligns human-robot visual features via semantic distribution matching on corresponding body parts plus Gram loss, yielding 65% better zero-shot policy transfer than baseline DINO.
Contact-Grounded Policy predicts coupled robot-state and tactile trajectories with a diffusion model and maps them via a learned consistency function to executable targets for compliance controllers, outperforming standard visuotactile diffusion baselines on physical and simulated dexterous tasks.
Self-supervised fingertip mapping with few-shot human anchors and a pinch contact classifier yields calibration-free, more intuitive retargeting across diverse human-like robot hands.
A JAX-implemented flow-based equivariant model for multi-embodiment grasping that deduces kinematics from geometry to support variable-DoF grippers with a new dataset of 25k scenes and 20M grasps.
DexTeleop-0 adds a tactile-driven adaptation loop to bimanual dexterous teleoperation that estimates contact points and applies localized force-compliant corrections via operational-space Jacobian updates.
GraspSense computes force maps from object geometry to select mechanically safe grasp regions and regulate grip forces for dexterous hands.
citing papers explorer
-
Human Universal Grasping
HUG trains a flow-matching model on a new 1M-frame egocentric human grasp dataset to generate retargetable grasps from single RGB-D images, beating baselines by 23-34% on a new 90-object benchmark.
-
Mobile UMI: Cross-View Diffusion Policy with Decoupled Kinematics for Mobile Manipulation
A hardware-free dual-camera capture framework with ChArUco spatial unification and receding-horizon state alignment enables decoupled SE(3) manipulation and SE(2) base trajectories for diffusion policies, yielding 83.8% average success on four long-horizon household tasks.
-
BiDexGrasp: Coordinated Bimanual Dexterous Grasps across Object Geometries and Sizes
BiDexGrasp supplies a 9.7-million-grasp bimanual dexterous dataset built via two-stage synthesis and a coordinated geometry-size-adaptive model that generates grasps for unseen objects.
-
LACE: Latent Visual Representation for Cross-Embodiment Learning
LACE aligns human-robot visual features via semantic distribution matching on corresponding body parts plus Gram loss, yielding 65% better zero-shot policy transfer than baseline DINO.
-
Contact-Grounded Policy: Dexterous Visuotactile Policy with Generative Contact Grounding
Contact-Grounded Policy predicts coupled robot-state and tactile trajectories with a diffusion model and maps them via a learned consistency function to executable targets for compliance controllers, outperforming standard visuotactile diffusion baselines on physical and simulated dexterous tasks.
-
AnyDexRT: Calibration-Free Dexterous Hand Retargeting with Few-Shot Human Guidance
Self-supervised fingertip mapping with few-shot human anchors and a pinch contact classifier yields calibration-free, more intuitive retargeting across diverse human-like robot hands.
-
Towards a Multi-Embodied Grasping Agent
A JAX-implemented flow-based equivariant model for multi-embodiment grasping that deduces kinematics from geometry to support variable-DoF grippers with a new dataset of 25k scenes and 20M grasps.
-
DexTeleop-0: Force-Aware Bimanual Dexterous Teleoperation with Ego-Centric Perception towards Shared Autonomy
DexTeleop-0 adds a tactile-driven adaptation loop to bimanual dexterous teleoperation that estimates contact points and applies localized force-compliant corrections via operational-space Jacobian updates.
-
GraspSense: Physically Grounded Grasp and Grip Planning for a Dexterous Robotic Hand via Language-Guided Perception and Force Maps
GraspSense computes force maps from object geometry to select mechanically safe grasp regions and regulate grip forces for dexterous hands.