Spatial-temporal fusion for high accuracy depth maps using dynamic MRFs

Jiejie Zhu; Liang Wang; Jizhou Gao; Ruigang Yang

doi:10.1109/TPAMI.2009.68

Spatial-temporal fusion for high accuracy depth maps using dynamic MRFs

Jiejie Zhu, Liang Wang, Jizhou Gao, Ruigang Yang

Computer Science

Research output: Contribution to journal › Article › peer-review

108 Scopus citations

Abstract

Time-of-flight range sensors and passive stereo have complimentary characteristics in nature. To fuse them to get high accuracy depth maps varying over time, we extend traditional spatial MRFs to dynamic MRFs with temporal coherence. This new model allows both the spatial and the temporal relationship to be propagated in local neighbors. By efficiently finding a maximum of the posterior probability using Loopy Belief Propagation, we show that our approach leads to improved accuracy and robustness of depth estimates for dynamic scenes.

Original language	English
Article number	4815256
Pages (from-to)	899-909
Number of pages	11
Journal	IEEE Transactions on Pattern Analysis and Machine Intelligence
Volume	32
Issue number	5
DOIs	https://doi.org/10.1109/TPAMI.2009.68
State	Published - 2010

Bibliographical note

Funding Information:
This work is supported in part by the University of Kentucky Research Foundation, the US Department of Homeland Security, US National Science Foundation Grants HCC-0448185 and CPA-0811647, and the Open Project Program of the State Key Lab of CAD&CG (grant no.: A0812), Zhejiang University, China. This work was performed while Jiejie Zhu was with the University of Kentucky.

Funding

This work is supported in part by the University of Kentucky Research Foundation, the US Department of Homeland Security, US National Science Foundation Grants HCC-0448185 and CPA-0811647, and the Open Project Program of the State Key Lab of CAD&CG (grant no.: A0812), Zhejiang University, China. This work was performed while Jiejie Zhu was with the University of Kentucky.

Funders	Funder number
US National Science Foundation	HCC-0448185, CPA-0811647
University of Kentucky and São Paulo Research Foundation
U.S. Department of Homeland Security
Zhejiang University
State Key Laboratory of Computer Aided Design and Computer Graphics	A0812

Keywords

Data fusion
Global optimization
MRFs
Stereo
Time-of-flight sensor

ASJC Scopus subject areas

Software
Computer Vision and Pattern Recognition
Computational Theory and Mathematics
Artificial Intelligence
Applied Mathematics

Access to Document

10.1109/TPAMI.2009.68

Cite this

@article{d0ef2cf9958f42c18e8a9ededc29ef61,

title = "Spatial-temporal fusion for high accuracy depth maps using dynamic MRFs",

abstract = "Time-of-flight range sensors and passive stereo have complimentary characteristics in nature. To fuse them to get high accuracy depth maps varying over time, we extend traditional spatial MRFs to dynamic MRFs with temporal coherence. This new model allows both the spatial and the temporal relationship to be propagated in local neighbors. By efficiently finding a maximum of the posterior probability using Loopy Belief Propagation, we show that our approach leads to improved accuracy and robustness of depth estimates for dynamic scenes.",

keywords = "Data fusion, Global optimization, MRFs, Stereo, Time-of-flight sensor",

author = "Jiejie Zhu and Liang Wang and Jizhou Gao and Ruigang Yang",

note = "Funding Information: This work is supported in part by the University of Kentucky Research Foundation, the US Department of Homeland Security, US National Science Foundation Grants HCC-0448185 and CPA-0811647, and the Open Project Program of the State Key Lab of CAD&CG (grant no.: A0812), Zhejiang University, China. This work was performed while Jiejie Zhu was with the University of Kentucky.",

year = "2010",

doi = "10.1109/TPAMI.2009.68",

language = "English",

volume = "32",

pages = "899--909",

number = "5",

}

TY - JOUR

T1 - Spatial-temporal fusion for high accuracy depth maps using dynamic MRFs

AU - Zhu, Jiejie

AU - Wang, Liang

AU - Gao, Jizhou

AU - Yang, Ruigang

N1 - Funding Information: This work is supported in part by the University of Kentucky Research Foundation, the US Department of Homeland Security, US National Science Foundation Grants HCC-0448185 and CPA-0811647, and the Open Project Program of the State Key Lab of CAD&CG (grant no.: A0812), Zhejiang University, China. This work was performed while Jiejie Zhu was with the University of Kentucky.

PY - 2010

Y1 - 2010

N2 - Time-of-flight range sensors and passive stereo have complimentary characteristics in nature. To fuse them to get high accuracy depth maps varying over time, we extend traditional spatial MRFs to dynamic MRFs with temporal coherence. This new model allows both the spatial and the temporal relationship to be propagated in local neighbors. By efficiently finding a maximum of the posterior probability using Loopy Belief Propagation, we show that our approach leads to improved accuracy and robustness of depth estimates for dynamic scenes.

AB - Time-of-flight range sensors and passive stereo have complimentary characteristics in nature. To fuse them to get high accuracy depth maps varying over time, we extend traditional spatial MRFs to dynamic MRFs with temporal coherence. This new model allows both the spatial and the temporal relationship to be propagated in local neighbors. By efficiently finding a maximum of the posterior probability using Loopy Belief Propagation, we show that our approach leads to improved accuracy and robustness of depth estimates for dynamic scenes.

KW - Data fusion

KW - Global optimization

KW - MRFs

KW - Stereo

KW - Time-of-flight sensor

UR - http://www.scopus.com/inward/record.url?scp=77949873218&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949873218&partnerID=8YFLogxK

U2 - 10.1109/TPAMI.2009.68

DO - 10.1109/TPAMI.2009.68

M3 - Article

C2 - 20299713

AN - SCOPUS:77949873218

SN - 0162-8828

VL - 32

SP - 899

EP - 909

JO - IEEE Transactions on Pattern Analysis and Machine Intelligence

JF - IEEE Transactions on Pattern Analysis and Machine Intelligence

IS - 5

M1 - 4815256

ER -

Spatial-temporal fusion for high accuracy depth maps using dynamic MRFs

Abstract

Bibliographical note

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this