arrow
返回

Scene Reconstruction with Functional Objects for Robot Autonomy

delete2022-09-20
delete12
PRE
AI
M
Muzhi Han
Z
Zeyu Zhang
Z
Ziyuan Jiao
X
Xu Xie
Y
Yixin Zhu *
Song-Chun Zhu 封面图
Song-Chun Zhu (Song‐Chun Zhu)
H
Hangxin Liu *
DOI:10.1007/s11263-022-01670-0delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
In this paper, we rethink the problem of scene reconstruction from an embodied agent's perspective: While the classic view focuses on the reconstruction accuracy, our new perspective emphasizes the underlying functions and constraints of the reconstructed scenes that provide actionable information for simulating interactions with agents. Here, we address this challenging problem by reconstructing a functionally equivalent and interactive scene from RGB-D data streams, where the objects within are segmented by a dedicated 3D volumetric panoptic mapping module and subsequently replaced by part-based articulated CAD models to afford finer-grained robot interactions. The object functionality and contextual relations are further organized by a graph-based scene representation that can be readily incorporated into robots' action specifications and task definition, facilitating their long-term task and motion planning in the scenes. In the experiments, we demonstrate that (i) our panoptic mapping module outperforms previous state-of-the-art methods in recognizing and segmenting scene entities, (ii) the geometric and physical reasoning procedure matches, aligns, and replaces object meshes with best-fitted CAD models, and (iii) the reconstructed functionally equivalent and interactive scenes are physically plausible and naturally afford actionable interactions; without any manual labeling, they are seamlessly imported to ROS-based robot simulators and VR environments for simulating complex robot interactions.
Keyword:
Functional scene representation
3D scene reconstruction
Actionable information
Volumetric panoptic mapping
Physical reasoning
Robot interaction

期刊

International Journal of Computer Vision 封面图
International Journal of Computer Vision
IF:
9.3
论文数:
3.9K
被引数:
2.8W

机构

U
university of california los angeles
学者数:
5.3W
论文数: 4.2W
被引数: 89
P
peking university
学者数:
11.9W
论文数: 8.7W
被引数: 146
University of California System 封面图
University of California System
学者数:
37.7W
论文数: 33.8W
被引数: 6.6K
学者 查看更多机构
引用论文

引用论文

In SituRhamnolipid Production at an Abandoned Petroleum Refinery
err2010-06-24
err0
PREAI
errDaniel P. Cassidy; Andrew J. Hudak; D. Dale Werkema; Estella A. Atekwana; Silvia Rossbach; Joseph W. Duris; Eliot A. Atekwana; and William A. Sauck
err分享
err收藏
err分享
err收藏
Cryo-EM reveals structural breaks in a patient-derived amyloid fibril from systemic AL amyloidosisCryo-em揭示了系统性AL淀粉样变性患者衍生的淀粉样蛋白原纤维的结构断裂
err
IF0
err2020-10-12
err0
errOAAI
errLynn Radamaker; Julian Baur; Stefanie Huhn; Christian Haupt; Ute Hegenbart; Stefan Schönland; Akanksha Bansal; Matthias Schmidt; Marcus Fändrich
err分享
err收藏
err分享
err收藏
XRD and ATR-FTIR techniques for integrity assessment of gamma radiation sterilized cortical bone pretreated by antioxidants
err2020-11-09
err0
PREAI
errNaglaa S. El-Hansi; Hoda H. Said; Omar S. Desouky; Mahmoud A. Khalaf; Mona S. Talaat; Abdelsattar M. Sallam
err分享
err收藏
Phytochemical and antiinflammatory studies on Terminalia catappa
err2004-06-01
err0
PREAI
errY.M. Fan; L.Z. Xu; J. Gao; Y. Wang; X.H. Tang; X.N. Zhao; Z.X. Zhang
err分享
err收藏
A tale of two explanations: Enhancing human trust by explaining robot behavior
err2019-12-18
err89
PREAI
errEdmonds, Mark; Gao, Feng; Liu, Hangxin; Xie, Xu; Qi, Siyuan; Rothrock, Brandon; Zhu, Yixin; Wu, Ying Nian; Lu, Hongjing; Zhu, Song-Chun
err分享
err收藏
学者 查看更多内容