EmbodMocap: In-the-Wild 4D Human-Scene Reconstruction for Embodied Agents

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Wenjia, Pan, Liang, Pi, Huaijin, Lou, Yuke, Ren, Xuqian, Wu, Yifan, Liao, Zhouyingcheng, Yang, Lei, Dabral, Rishabh, Theobalt, Christian, Komura, Taku
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866912997395398656
author Wang, Wenjia
Pan, Liang
Pi, Huaijin
Lou, Yuke
Ren, Xuqian
Wu, Yifan
Liao, Zhouyingcheng
Yang, Lei
Dabral, Rishabh
Theobalt, Christian
Komura, Taku
author_facet Wang, Wenjia
Pan, Liang
Pi, Huaijin
Lou, Yuke
Ren, Xuqian
Wu, Yifan
Liao, Zhouyingcheng
Yang, Lei
Dabral, Rishabh
Theobalt, Christian
Komura, Taku
contents Human behaviors in the real world naturally encode rich, long-term contextual information that can be leveraged to train embodied agents for perception, understanding, and acting. However, existing capture systems typically rely on costly studio setups and wearable devices, limiting the large-scale collection of scene-conditioned human motion data in the wild. To address this, we propose EmbodMocap, a portable and affordable data collection pipeline using two moving iPhones. Our key idea is to jointly calibrate dual RGB-D sequences to reconstruct both humans and scenes within a unified metric world coordinate frame. The proposed method allows metric-scale and scene-consistent capture in everyday environments without static cameras or markers, bridging human motion and scene geometry seamlessly. Compared with optical capture ground truth, we demonstrate that the dual-view setting exhibits a remarkable ability to mitigate depth ambiguity, achieving superior alignment and reconstruction performance over single iphone or monocular models. Based on the collected data, we empower three embodied AI tasks: monocular human-scene-reconstruction, where we fine-tune on feedforward models that output metric-scale, world-space aligned humans and scenes; physics-based character animation, where we prove our data could be used to scale human-object interaction skills and scene-aware motion tracking; and robot motion control, where we train a humanoid robot via sim-to-real RL to replicate human motions depicted in videos. Experimental results validate the effectiveness of our pipeline and its contributions towards advancing embodied AI research.
format Preprint
id arxiv_https___arxiv_org_abs_2602_23205
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EmbodMocap: In-the-Wild 4D Human-Scene Reconstruction for Embodied Agents
Wang, Wenjia
Pan, Liang
Pi, Huaijin
Lou, Yuke
Ren, Xuqian
Wu, Yifan
Liao, Zhouyingcheng
Yang, Lei
Dabral, Rishabh
Theobalt, Christian
Komura, Taku
Computer Vision and Pattern Recognition
Human behaviors in the real world naturally encode rich, long-term contextual information that can be leveraged to train embodied agents for perception, understanding, and acting. However, existing capture systems typically rely on costly studio setups and wearable devices, limiting the large-scale collection of scene-conditioned human motion data in the wild. To address this, we propose EmbodMocap, a portable and affordable data collection pipeline using two moving iPhones. Our key idea is to jointly calibrate dual RGB-D sequences to reconstruct both humans and scenes within a unified metric world coordinate frame. The proposed method allows metric-scale and scene-consistent capture in everyday environments without static cameras or markers, bridging human motion and scene geometry seamlessly. Compared with optical capture ground truth, we demonstrate that the dual-view setting exhibits a remarkable ability to mitigate depth ambiguity, achieving superior alignment and reconstruction performance over single iphone or monocular models. Based on the collected data, we empower three embodied AI tasks: monocular human-scene-reconstruction, where we fine-tune on feedforward models that output metric-scale, world-space aligned humans and scenes; physics-based character animation, where we prove our data could be used to scale human-object interaction skills and scene-aware motion tracking; and robot motion control, where we train a humanoid robot via sim-to-real RL to replicate human motions depicted in videos. Experimental results validate the effectiveness of our pipeline and its contributions towards advancing embodied AI research.
title EmbodMocap: In-the-Wild 4D Human-Scene Reconstruction for Embodied Agents
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.23205