Pith. sign in

REVIEW 5 cited by

Crossing the Human-Robot Embodiment Gap with Sim-to-Real RL using One Human Demonstration

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2504.12609 v3 pith:MITFM7XA submitted 2025-04-17 cs.RO cs.AI

classification cs.ROcs.AI
keywords learningembodimenthumanhuman2sim2robotmanipulationdexteroushuman-robotpose
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Teaching robots dexterous manipulation skills often requires collecting hundreds of demonstrations using wearables or teleoperation, a process that is challenging to scale. Videos of human-object interactions are easier to collect and scale, but leveraging them directly for robot learning is difficult due to the lack of explicit action labels and human-robot embodiment differences. We propose Human2Sim2Robot, a novel real-to-sim-to-real framework for training dexterous manipulation policies using only one RGB-D video of a human demonstrating a task. Our method utilizes reinforcement learning (RL) in simulation to cross the embodiment gap without relying on wearables, teleoperation, or large-scale data collection. From the video, we extract: (1) the object pose trajectory to define an object-centric, embodiment-agnostic reward, and (2) the pre-manipulation hand pose to initialize and guide exploration during RL training. These components enable effective policy learning without any task-specific reward tuning. In the single human demo regime, Human2Sim2Robot outperforms object-aware replay by over 55% and imitation learning by over 68% on grasping, non-prehensile manipulation, and multi-step tasks. Website: https://human2sim2robot.github.io

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. DemoBridge: A Simulation-in-the-Loop Toolkit for Single-View Human Demonstration Retargeting

    cs.RO 2026-07 conditional novelty 6.0 of 10

    DemoBridge retargets single-view human hand demonstrations into physics-validated, collision-aware robot trajectories via whole-trajectory optimization and simulation-in-the-loop re-planning.

  2. Play2Perfect: What Matters in Dexterous Play Pretraining for Precise Assembly?

    cs.RO 2026-06 unverdicted novelty 6.0 of 10

    Task-agnostic RL play pretraining on diverse objects yields a reusable dexterous prior that makes sparse-reward assembly learning ~33× more sample-efficient and enables zero-shot sim-to-real transfer on tight insertio...

  3. Grasp to Act: Dexterous Grasping for Tool Use in Dynamic Settings

    cs.RO 2026-02 conditional novelty 6.0 of 10

    Combining wrench-tested grasp optimization with real-time RL finger adjustments lets a 16-DoF robot hand keep tools stable during hammering, sawing, cutting, stirring, and scooping.

  4. Masquerade: Learning from In-the-wild Human Videos using Data-Editing

    cs.RO 2025-08 conditional novelty 6.0 of 10

    Editing in-the-wild egocentric human videos into robot-overlaid clips and co-training a vision encoder on them improves zero-shot generalization of bimanual policies trained from only 50 demonstrations per task.

  5. HERMES: Human-to-Robot Embodied Learning from Multi-Source Motion Data for Mobile Dexterous Manipulation

    cs.RO 2025-08 conditional novelty 5.0 of 10

    HERMES converts a single human motion demonstration into a deployable mobile bimanual dexterous manipulation policy, using RL, depth-image distillation, and closed-loop PnP pose refinement.

Pith tools