EgoHumanoid: Unlocking In-the-Wild Loco-Manipulation with Robot-Free Egocentric Demonstration

2026-02-10

AI Summary

This work addresses the limited generalization of mobile manipulation in humanoid robots operating in real-world complex environments, primarily due to scarce robot-collected data. To overcome this challenge, the authors propose a cross-domain policy learning framework that integrates large-scale first-person human demonstration videos with a small amount of robot data. By establishing a systematic human-to-humanoid alignment mechanism that does not require robot participation—leveraging viewpoint alignment, unified action space mapping, and joint vision-language-action training—the approach effectively bridges morphological and perspective discrepancies. Experimental results demonstrate a 51% performance improvement over baselines trained solely on robot data in real-world settings, with significantly enhanced generalization to unseen scenarios. These findings validate the scalability and efficacy of leveraging abundant human demonstration data for skill transfer to humanoid robots.

Technology Category

• Intelligent Robots: Manipulation
• Humans and AI: Learning Human Values and Preferences
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

Human demonstrations offer rich environmental diversity and scale naturally, making them an appealing alternative to robot teleoperation. While this paradigm has advanced robot-arm manipulation, its potential for the more challenging, data-hungry problem of humanoid loco-manipulation remains largely unexplored. We present EgoHumanoid, the first framework to co-train a vision-language-action policy using abundant egocentric human demonstrations together with a limited amount of robot data, enabling humanoids to perform loco-manipulation across diverse real-world environments. To bridge the embodiment gap between humans and robots, including discrepancies in physical morphology and viewpoint, we introduce a systematic alignment pipeline spanning from hardware design to data processing. A portable system for scalable human data collection is developed, and we establish practical collection protocols to improve transferability. At the core of our human-to-humanoid alignment pipeline lies two key components. The view alignment reduces visual domain discrepancies caused by camera height and perspective variation. The action alignment maps human motions into a unified, kinematically feasible action space for humanoid control. Extensive real-world experiments demonstrate that incorporating robot-free egocentric data significantly outperforms robot-only baselines by 51\%, particularly in unseen environments. Our analysis further reveals which behaviors transfer effectively and the potential for scaling human data.

Problem

• loco-manipulation
• egocentric demonstration
• humanoid robot
• embodiment gap
• in-the-wild

Innovation

• egocentric demonstration
• humanoid locomotion
• view alignment
• action alignment
• vision-language-action policy