LIDEA: Human-to-Robot Imitation Learning via Implicit Feature Distillation and Explicit Geometry Alignment

TL;DR

LIDEA is a novel imitation learning framework that leverages human videos to improve robot learning by aligning features and geometry across embodiments, enhancing data efficiency and robustness.

Contribution

LIDEA introduces a dual-stage transitive distillation and embodiment-agnostic geometric alignment to effectively transfer human demonstrations to robots.

Findings

Human data can replace up to 80% of robot demonstrations.

LIDEA improves out-of-distribution generalization.

Framework enhances data efficiency and robustness in robot learning.

Abstract

Scaling up robot learning is hindered by the scarcity of robotic demonstrations, whereas human videos offer a vast, untapped source of interaction data. However, bridging the embodiment gap between human hands and robot arms remains a critical challenge. Existing cross-embodiment transfer strategies typically rely on visual editing, but they often introduce visual artifacts due to intrinsic discrepancies in visual appearance and 3D geometry. To address these limitations, we introduce LIDEA (Implicit Feature Distillation and Explicit Geometric Alignment), an imitation learning framework in which policy learning benefits from human demonstrations. In the 2D visual domain, LIDEA employs a dual-stage transitive distillation pipeline that aligns human and robot representations in a shared latent space. In the 3D geometric domain, we propose an embodiment-agnostic alignment strategy that explicitly decouples embodiment from interaction geometry, ensuring consistent 3D-aware perception. Extensive experiments empirically validate LIDEA from two perspectives: data efficiency and OOD robustness. Results show that human data substitutes up to 80% of costly robot demonstrations, and the framework successfully transfers unseen patterns from human videos for out-of-distribution generalization.