EgoZero: Robot Learning from Smart Glasses

TL;DR

EgoZero leverages egocentric human demonstrations captured with smart glasses to train robot manipulation policies without robot data, achieving high success rates in real-world tasks with minimal data.

Contribution

This work introduces EgoZero, a system that learns robot manipulation policies from human egocentric data without requiring robot demonstrations, enabling scalable and generalizable robot learning.

Findings

70% success rate over 7 tasks with 20 minutes of data per task

Zero-shot transfer of policies to a physical robot

Effective extraction of robot actions from human egocentric videos

Abstract

Despite recent progress in general purpose robotics, robot policies still lag far behind basic human capabilities in the real world. Humans interact constantly with the physical world, yet this rich data resource remains largely untapped in robot learning. We propose EgoZero, a minimal system that learns robust manipulation policies from human demonstrations captured with Project Aria smart glasses, $\textbf{and zero robot data}$. EgoZero enables: (1) extraction of complete, robot-executable actions from in-the-wild, egocentric, human demonstrations, (2) compression of human visual observations into morphology-agnostic state representations, and (3) closed-loop policy learning that generalizes morphologically, spatially, and semantically. We deploy EgoZero policies on a gripper Franka Panda robot and demonstrate zero-shot transfer with 70% success rate over 7 manipulation tasks and only 20 minutes of data collection per task. Our results suggest that in-the-wild human data can serve as a scalable foundation for real-world robot learning - paving the way toward a future of abundant, diverse, and naturalistic training data for robots. Code and videos are available at https://egozero-robot.github.io.