Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning

Bryan P. Maldonado; Nolan W. Hayes; Diana Hun

doi:10.22260/ISARC2023/0009

Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning

Bryan P. Maldonado, Nolan W. Hayes, Diana Hun

Building Envelope Materials Research

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

Modern retrofit construction practices use 3D point cloud data of the building envelope to obtain the as-built dimensions. However, manual segmentation by a trained professional is required to identify and measure window openings, door openings, and other architectural features, making the use of 3D point clouds labor-intensive. In this study, the Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) algorithm is described, which can significantly reduce the time spent during point cloud segmentation. The Auto-CuBES algorithm inputs a 3D point cloud generated by commonly available surveying equipment and outputs a wire-frame model of the building envelope. Unsupervised machine learning methods were used to identify facades, windows, and doors while minimizing the number of calibration parameters. Additionally, Auto-CuBES generates a heat map of each facade indicating non-planar characteristics that are crucial for the optimization of connections used in overclad envelope retrofits. With a scan resolution of 3 mm, the resulting window dimensions showed a mean absolute error of 4.2 mm compared to manual laser measurements.

Original language	English
Title of host publication	Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023
Editors	Borja Garcia de Soto, Vicente Gonzalez, Ioannis Brilakis
Publisher	International Association for Automation and Robotics in Construction (IAARC)
Pages	48-55
Number of pages	8
ISBN (Electronic)	9780645832204
DOIs	https://doi.org/10.22260/ISARC2023/0009
State	Published - 2023
Event	40th International Symposium on Automation and Robotics in Construction, ISARC 2023 - Chennai, India Duration: Jul 5 2023 → Jul 7 2023

Publication series

Name	Proceedings of the International Symposium on Automation and Robotics in Construction
ISSN (Electronic)	2413-5844

Conference

Conference	40th International Symposium on Automation and Robotics in Construction, ISARC 2023
Country/Territory	India
City	Chennai
Period	07/5/23 → 07/7/23

Funding

∗This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The publisher acknowledges the US government license to provide public access under the DOE Public Access Plan (https://energy.gov/downloads/doe-public-access-plan). This research was supported by the DOE Office of Energy Efficiency and Renewable Energy (EERE), Building Technologies Office, under contract DE-AC05-00OR22725, and used resources at the Building Technologies Research and Integration Center, a DOE-EERE User Facility at Oak Ridge National Laboratory. This manuscript has been authored by UT-Battelle, LLC, under contract DE-AC05-00OR22725 with the US Department of Energy (DOE). The publisher acknowledges the US government license to provide public access under the DOE Public Access Plan (https://energy.gov/downloads/doe-public-access-plan).

Keywords

point cloud
segmentation
unsupervised learning

Access to Document

10.22260/ISARC2023/0009

Cite this

Maldonado, B. P., Hayes, N. W., & Hun, D. (2023). Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning. In B. Garcia de Soto, V. Gonzalez, & I. Brilakis (Eds.), Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023 (pp. 48-55). (Proceedings of the International Symposium on Automation and Robotics in Construction). International Association for Automation and Robotics in Construction (IAARC). https://doi.org/10.22260/ISARC2023/0009

Maldonado, Bryan P. ; Hayes, Nolan W. ; Hun, Diana. / Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning. Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023. editor / Borja Garcia de Soto ; Vicente Gonzalez ; Ioannis Brilakis. International Association for Automation and Robotics in Construction (IAARC), 2023. pp. 48-55 (Proceedings of the International Symposium on Automation and Robotics in Construction).

@inproceedings{53384c75a4ca43b2857fd7f8eee4f67c,

title = "Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning",

abstract = "Modern retrofit construction practices use 3D point cloud data of the building envelope to obtain the as-built dimensions. However, manual segmentation by a trained professional is required to identify and measure window openings, door openings, and other architectural features, making the use of 3D point clouds labor-intensive. In this study, the Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) algorithm is described, which can significantly reduce the time spent during point cloud segmentation. The Auto-CuBES algorithm inputs a 3D point cloud generated by commonly available surveying equipment and outputs a wire-frame model of the building envelope. Unsupervised machine learning methods were used to identify facades, windows, and doors while minimizing the number of calibration parameters. Additionally, Auto-CuBES generates a heat map of each facade indicating non-planar characteristics that are crucial for the optimization of connections used in overclad envelope retrofits. With a scan resolution of 3 mm, the resulting window dimensions showed a mean absolute error of 4.2 mm compared to manual laser measurements.",

keywords = "point cloud, segmentation, unsupervised learning",

author = "Maldonado, {Bryan P.} and Hayes, {Nolan W.} and Diana Hun",

note = "Publisher Copyright: {\textcopyright} ISARC 2023. All rights reserved.; 40th International Symposium on Automation and Robotics in Construction, ISARC 2023 ; Conference date: 05-07-2023 Through 07-07-2023",

year = "2023",

doi = "10.22260/ISARC2023/0009",

language = "English",

series = "Proceedings of the International Symposium on Automation and Robotics in Construction",

publisher = "International Association for Automation and Robotics in Construction (IAARC)",

pages = "48--55",

editor = "{Garcia de Soto}, Borja and Vicente Gonzalez and Ioannis Brilakis",

booktitle = "Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023",

}

Maldonado, BP , Hayes, NW & Hun, D 2023, Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning. in B Garcia de Soto, V Gonzalez & I Brilakis (eds), Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023. Proceedings of the International Symposium on Automation and Robotics in Construction, International Association for Automation and Robotics in Construction (IAARC), pp. 48-55, 40th International Symposium on Automation and Robotics in Construction, ISARC 2023, Chennai, India, 07/5/23. https://doi.org/10.22260/ISARC2023/0009

Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning. / Maldonado, Bryan P.; Hayes, Nolan W.; Hun, Diana.
Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023. ed. / Borja Garcia de Soto; Vicente Gonzalez; Ioannis Brilakis. International Association for Automation and Robotics in Construction (IAARC), 2023. p. 48-55 (Proceedings of the International Symposium on Automation and Robotics in Construction).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning

AU - Maldonado, Bryan P.

AU - Hayes, Nolan W.

AU - Hun, Diana

PY - 2023

Y1 - 2023

N2 - Modern retrofit construction practices use 3D point cloud data of the building envelope to obtain the as-built dimensions. However, manual segmentation by a trained professional is required to identify and measure window openings, door openings, and other architectural features, making the use of 3D point clouds labor-intensive. In this study, the Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) algorithm is described, which can significantly reduce the time spent during point cloud segmentation. The Auto-CuBES algorithm inputs a 3D point cloud generated by commonly available surveying equipment and outputs a wire-frame model of the building envelope. Unsupervised machine learning methods were used to identify facades, windows, and doors while minimizing the number of calibration parameters. Additionally, Auto-CuBES generates a heat map of each facade indicating non-planar characteristics that are crucial for the optimization of connections used in overclad envelope retrofits. With a scan resolution of 3 mm, the resulting window dimensions showed a mean absolute error of 4.2 mm compared to manual laser measurements.

AB - Modern retrofit construction practices use 3D point cloud data of the building envelope to obtain the as-built dimensions. However, manual segmentation by a trained professional is required to identify and measure window openings, door openings, and other architectural features, making the use of 3D point clouds labor-intensive. In this study, the Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) algorithm is described, which can significantly reduce the time spent during point cloud segmentation. The Auto-CuBES algorithm inputs a 3D point cloud generated by commonly available surveying equipment and outputs a wire-frame model of the building envelope. Unsupervised machine learning methods were used to identify facades, windows, and doors while minimizing the number of calibration parameters. Additionally, Auto-CuBES generates a heat map of each facade indicating non-planar characteristics that are crucial for the optimization of connections used in overclad envelope retrofits. With a scan resolution of 3 mm, the resulting window dimensions showed a mean absolute error of 4.2 mm compared to manual laser measurements.

KW - point cloud

KW - segmentation

KW - unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=85172907070&partnerID=8YFLogxK

U2 - 10.22260/ISARC2023/0009

DO - 10.22260/ISARC2023/0009

M3 - Conference contribution

AN - SCOPUS:85172907070

T3 - Proceedings of the International Symposium on Automation and Robotics in Construction

SP - 48

EP - 55

BT - Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023

A2 - Garcia de Soto, Borja

A2 - Gonzalez, Vicente

A2 - Brilakis, Ioannis

PB - International Association for Automation and Robotics in Construction (IAARC)

T2 - 40th International Symposium on Automation and Robotics in Construction, ISARC 2023

Y2 - 5 July 2023 through 7 July 2023

ER -

Maldonado BP , Hayes NW , Hun D. Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning. In Garcia de Soto B, Gonzalez V, Brilakis I, editors, Proceedings of the 40th International Symposium on Automation and Robotics in Construction, ISARC 2023. International Association for Automation and Robotics in Construction (IAARC). 2023. p. 48-55. (Proceedings of the International Symposium on Automation and Robotics in Construction). doi: 10.22260/ISARC2023/0009

Automatic point Cloud Building Envelope Segmentation (Auto-CuBES) using Machine Learning

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this