Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation

Chen Feng, Vineet R. Kamat, Carol C. Menassa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Accurately modeling as-built environments and tracking moving objects' poses are critical for many architecture, engineering, construction, and facility management (AECFM) automation applications. Equally important are the reliability, operating range and cost efficiency of such solutions for their broad deployment in unstructured, dynamic, and sometimes featureless AECFM sites. In this paper, a flexible vision-based technique is developed for accurate, robust, low-cost, and scalable pose estimation and as-built modeling in AECFM applications. This technique combines marker-based pose estimation and structure-from-motion (SfM). In the preparation phase, a sparse set of visual markers are installed in the target environment. During the operation phase, a set of unordered images are taken with a calibrated RGB camera. These images are immediately processed by a SfM system to estimate those markers' poses and generate a sparse point cloud, which can be used by robots or other mobile clients for either moving objects' pose estimation, or dimensional analysis of that environment. Furthermore, for as-built modeling, the RGB camera is replaced by a RGBD camera to create both a dense 3D point cloud and a concise planar model of the environment. Experiments have demonstrated sufficient accuracy (average absolute error within 5 mm over a 9 m scale) of the proposed technique.

Original languageEnglish (US)
Title of host publicationConstruction Research Congress 2016
Subtitle of host publicationOld and New Construction Technologies Converge in Historic San Juan - Proceedings of the 2016 Construction Research Congress, CRC 2016
EditorsJose L. Perdomo-Rivera, Carla Lopez del Puerto, Antonio Gonzalez-Quevedo, Francisco Maldonado-Fortunet, Omar I. Molina-Bas
PublisherAmerican Society of Civil Engineers (ASCE)
Pages2604-2613
Number of pages10
ISBN (Electronic)9780784479827
DOIs
StatePublished - Jan 1 2016
EventConstruction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan, CRC 2016 - San Juan, Puerto Rico
Duration: May 31 2016Jun 2 2016

Other

OtherConstruction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan, CRC 2016
CountryPuerto Rico
CitySan Juan
Period5/31/166/2/16

Fingerprint

Cameras
Costs
Automation
Robots
Experiments

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Building and Construction

Cite this

Feng, C., Kamat, V. R., & Menassa, C. C. (2016). Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation. In J. L. Perdomo-Rivera, C. Lopez del Puerto, A. Gonzalez-Quevedo, F. Maldonado-Fortunet, & O. I. Molina-Bas (Eds.), Construction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan - Proceedings of the 2016 Construction Research Congress, CRC 2016 (pp. 2604-2613). American Society of Civil Engineers (ASCE). https://doi.org/10.1061/9780784479827.259

Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation. / Feng, Chen; Kamat, Vineet R.; Menassa, Carol C.

Construction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan - Proceedings of the 2016 Construction Research Congress, CRC 2016. ed. / Jose L. Perdomo-Rivera; Carla Lopez del Puerto; Antonio Gonzalez-Quevedo; Francisco Maldonado-Fortunet; Omar I. Molina-Bas. American Society of Civil Engineers (ASCE), 2016. p. 2604-2613.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Feng, C, Kamat, VR & Menassa, CC 2016, Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation. in JL Perdomo-Rivera, C Lopez del Puerto, A Gonzalez-Quevedo, F Maldonado-Fortunet & OI Molina-Bas (eds), Construction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan - Proceedings of the 2016 Construction Research Congress, CRC 2016. American Society of Civil Engineers (ASCE), pp. 2604-2613, Construction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan, CRC 2016, San Juan, Puerto Rico, 5/31/16. https://doi.org/10.1061/9780784479827.259
Feng C, Kamat VR, Menassa CC. Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation. In Perdomo-Rivera JL, Lopez del Puerto C, Gonzalez-Quevedo A, Maldonado-Fortunet F, Molina-Bas OI, editors, Construction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan - Proceedings of the 2016 Construction Research Congress, CRC 2016. American Society of Civil Engineers (ASCE). 2016. p. 2604-2613 https://doi.org/10.1061/9780784479827.259
Feng, Chen ; Kamat, Vineet R. ; Menassa, Carol C. / Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation. Construction Research Congress 2016: Old and New Construction Technologies Converge in Historic San Juan - Proceedings of the 2016 Construction Research Congress, CRC 2016. editor / Jose L. Perdomo-Rivera ; Carla Lopez del Puerto ; Antonio Gonzalez-Quevedo ; Francisco Maldonado-Fortunet ; Omar I. Molina-Bas. American Society of Civil Engineers (ASCE), 2016. pp. 2604-2613
@inproceedings{431d421d4074400f96d965d8938ee2c6,
title = "Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation",
abstract = "Accurately modeling as-built environments and tracking moving objects' poses are critical for many architecture, engineering, construction, and facility management (AECFM) automation applications. Equally important are the reliability, operating range and cost efficiency of such solutions for their broad deployment in unstructured, dynamic, and sometimes featureless AECFM sites. In this paper, a flexible vision-based technique is developed for accurate, robust, low-cost, and scalable pose estimation and as-built modeling in AECFM applications. This technique combines marker-based pose estimation and structure-from-motion (SfM). In the preparation phase, a sparse set of visual markers are installed in the target environment. During the operation phase, a set of unordered images are taken with a calibrated RGB camera. These images are immediately processed by a SfM system to estimate those markers' poses and generate a sparse point cloud, which can be used by robots or other mobile clients for either moving objects' pose estimation, or dimensional analysis of that environment. Furthermore, for as-built modeling, the RGB camera is replaced by a RGBD camera to create both a dense 3D point cloud and a concise planar model of the environment. Experiments have demonstrated sufficient accuracy (average absolute error within 5 mm over a 9 m scale) of the proposed technique.",
author = "Chen Feng and Kamat, {Vineet R.} and Menassa, {Carol C.}",
year = "2016",
month = "1",
day = "1",
doi = "10.1061/9780784479827.259",
language = "English (US)",
pages = "2604--2613",
editor = "Perdomo-Rivera, {Jose L.} and {Lopez del Puerto}, Carla and Antonio Gonzalez-Quevedo and Francisco Maldonado-Fortunet and Molina-Bas, {Omar I.}",
booktitle = "Construction Research Congress 2016",
publisher = "American Society of Civil Engineers (ASCE)",

}

TY - GEN

T1 - Marker-Assisted Structure from Motion for 3D Environment Modeling and Object Pose Estimation

AU - Feng, Chen

AU - Kamat, Vineet R.

AU - Menassa, Carol C.

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Accurately modeling as-built environments and tracking moving objects' poses are critical for many architecture, engineering, construction, and facility management (AECFM) automation applications. Equally important are the reliability, operating range and cost efficiency of such solutions for their broad deployment in unstructured, dynamic, and sometimes featureless AECFM sites. In this paper, a flexible vision-based technique is developed for accurate, robust, low-cost, and scalable pose estimation and as-built modeling in AECFM applications. This technique combines marker-based pose estimation and structure-from-motion (SfM). In the preparation phase, a sparse set of visual markers are installed in the target environment. During the operation phase, a set of unordered images are taken with a calibrated RGB camera. These images are immediately processed by a SfM system to estimate those markers' poses and generate a sparse point cloud, which can be used by robots or other mobile clients for either moving objects' pose estimation, or dimensional analysis of that environment. Furthermore, for as-built modeling, the RGB camera is replaced by a RGBD camera to create both a dense 3D point cloud and a concise planar model of the environment. Experiments have demonstrated sufficient accuracy (average absolute error within 5 mm over a 9 m scale) of the proposed technique.

AB - Accurately modeling as-built environments and tracking moving objects' poses are critical for many architecture, engineering, construction, and facility management (AECFM) automation applications. Equally important are the reliability, operating range and cost efficiency of such solutions for their broad deployment in unstructured, dynamic, and sometimes featureless AECFM sites. In this paper, a flexible vision-based technique is developed for accurate, robust, low-cost, and scalable pose estimation and as-built modeling in AECFM applications. This technique combines marker-based pose estimation and structure-from-motion (SfM). In the preparation phase, a sparse set of visual markers are installed in the target environment. During the operation phase, a set of unordered images are taken with a calibrated RGB camera. These images are immediately processed by a SfM system to estimate those markers' poses and generate a sparse point cloud, which can be used by robots or other mobile clients for either moving objects' pose estimation, or dimensional analysis of that environment. Furthermore, for as-built modeling, the RGB camera is replaced by a RGBD camera to create both a dense 3D point cloud and a concise planar model of the environment. Experiments have demonstrated sufficient accuracy (average absolute error within 5 mm over a 9 m scale) of the proposed technique.

UR - http://www.scopus.com/inward/record.url?scp=84976351087&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84976351087&partnerID=8YFLogxK

U2 - 10.1061/9780784479827.259

DO - 10.1061/9780784479827.259

M3 - Conference contribution

SP - 2604

EP - 2613

BT - Construction Research Congress 2016

A2 - Perdomo-Rivera, Jose L.

A2 - Lopez del Puerto, Carla

A2 - Gonzalez-Quevedo, Antonio

A2 - Maldonado-Fortunet, Francisco

A2 - Molina-Bas, Omar I.

PB - American Society of Civil Engineers (ASCE)

ER -