Saved in:
Bibliographic Details
Main Authors: Lin, Xiaopeng, Lian, Shijie, Yu, Bin, Yang, Ruoqi, Shen, Zhaolong, Wu, Changti, Miao, Yuzhuo, Jin, Yurun, Shi, Yukun, He, Jiyan, Huang, Cong, Cheng, Bojun, Chen, Kai
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.16793
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908812571574272
author Lin, Xiaopeng
Lian, Shijie
Yu, Bin
Yang, Ruoqi
Shen, Zhaolong
Wu, Changti
Miao, Yuzhuo
Jin, Yurun
Shi, Yukun
He, Jiyan
Huang, Cong
Cheng, Bojun
Chen, Kai
author_facet Lin, Xiaopeng
Lian, Shijie
Yu, Bin
Yang, Ruoqi
Shen, Zhaolong
Wu, Changti
Miao, Yuzhuo
Jin, Yurun
Shi, Yukun
He, Jiyan
Huang, Cong
Cheng, Bojun
Chen, Kai
contents Robotic generalization relies on physical intelligence: the ability to reason about state changes, contact-rich interactions, and long-horizon planning under egocentric perception and action. Vision Language Models (VLMs) are essential to Vision-Language-Action (VLA) systems, but the reliance on third-person training data creates a viewpoint gap for humanoid robots. Collecting massive robot-centric data is an ideal but impractical solution due to cost and diversity constraints. Conversely, human egocentric videos offer a highly scalable data source with rich interaction context, yet the embodiment mismatch prevents the direct application. To bridge this gap, we propose an Egocentric2Embodiment Translation Pipeline that transforms raw human egocentric videos into multi-level, schema-driven embodiment supervision with enforced evidence grounding and temporal consistency, enabling the construction of the Egocentric2Embodiment dataset (E2E-3M) at scale. An egocentric-aware embodied brain, termed PhysBrain, is obtained by training on the E2E-3M dataset. PhysBrain exhibits substantially improved egocentric understanding, particularly for planning. It provides an egocentric-aware initialization that enables more sample-efficient VLA fine-tuning and higher success rates, demonstrating effective transfer from human egocentric supervision to downstream robot control.
format Preprint
id arxiv_https___arxiv_org_abs_2512_16793
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PhysBrain: Human Egocentric Data as a Bridge from Vision Language Models to Physical Intelligence
Lin, Xiaopeng
Lian, Shijie
Yu, Bin
Yang, Ruoqi
Shen, Zhaolong
Wu, Changti
Miao, Yuzhuo
Jin, Yurun
Shi, Yukun
He, Jiyan
Huang, Cong
Cheng, Bojun
Chen, Kai
Robotics
Robotic generalization relies on physical intelligence: the ability to reason about state changes, contact-rich interactions, and long-horizon planning under egocentric perception and action. Vision Language Models (VLMs) are essential to Vision-Language-Action (VLA) systems, but the reliance on third-person training data creates a viewpoint gap for humanoid robots. Collecting massive robot-centric data is an ideal but impractical solution due to cost and diversity constraints. Conversely, human egocentric videos offer a highly scalable data source with rich interaction context, yet the embodiment mismatch prevents the direct application. To bridge this gap, we propose an Egocentric2Embodiment Translation Pipeline that transforms raw human egocentric videos into multi-level, schema-driven embodiment supervision with enforced evidence grounding and temporal consistency, enabling the construction of the Egocentric2Embodiment dataset (E2E-3M) at scale. An egocentric-aware embodied brain, termed PhysBrain, is obtained by training on the E2E-3M dataset. PhysBrain exhibits substantially improved egocentric understanding, particularly for planning. It provides an egocentric-aware initialization that enables more sample-efficient VLA fine-tuning and higher success rates, demonstrating effective transfer from human egocentric supervision to downstream robot control.
title PhysBrain: Human Egocentric Data as a Bridge from Vision Language Models to Physical Intelligence
topic Robotics
url https://arxiv.org/abs/2512.16793