Floor-plan Generation from Noisy Point Clouds

Xin Liu; Egor Bondarau; Peter H.N. de With

doi:10.1117/12.2679358

Floor-plan Generation from Noisy Point Clouds

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

61 Downloads (Pure)

Abstract

This paper proposes a growing-based floor-plan generation method that creates the global layout of buildings from noisy point clouds obtained by a stereo camera. We introduce a PCA-based line-growing concept with a subsequent filtering step, which is able to robustly handle the high noise levels in input point clouds. Experimental results show that this method outperforms the state-of-the-art techniques in floor-plan generation. The average score for building layouts has increased from 0.38 to 0.66 on our test dataset, compared to the previous best floor-plan generation method. Furthermore, the resulting floor plans are multiple thousands of times smaller in memory size than the input point clouds, while still preserving the main building structures.

Original language	English
Title of host publication	Fifteenth International Conference on Machine Vision (ICMV 2022)
Editors	Wolfgang Osten, Dmitry Nikolaev, Jianhong Zhou
Publisher	SPIE
Number of pages	10
ISBN (Electronic)	9781510666191
ISBN (Print)	9781510666184
DOIs	https://doi.org/10.1117/12.2679358
Publication status	Published - 7 Jun 2023
Event	15th International Conference on Machine Vision, ICMV 2022 - Rome, Italy Duration: 18 Nov 2022 → 20 Nov 2022 Conference number: 15 http://icmv.org/

Publication series

Name	Proceedings of SPIE
Volume	12701
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	15th International Conference on Machine Vision, ICMV 2022
Abbreviated title	ICMV 2022
Country/Territory	Italy
City	Rome
Period	18/11/22 → 20/11/22
Internet address	http://icmv.org/

Keywords

Floor-plan generation
indoor stereo SLAM
noisy point cloud filtering

Access to Document

10.1117/12.2679358

pdfFinal published version, 1.41 MBLicence: TAVERNE

Cite this

@inproceedings{f0bee9f655634321b8b07093363ce9d0,

title = "Floor-plan Generation from Noisy Point Clouds",

abstract = "This paper proposes a growing-based floor-plan generation method that creates the global layout of buildings from noisy point clouds obtained by a stereo camera. We introduce a PCA-based line-growing concept with a subsequent filtering step, which is able to robustly handle the high noise levels in input point clouds. Experimental results show that this method outperforms the state-of-the-art techniques in floor-plan generation. The average score for building layouts has increased from 0.38 to 0.66 on our test dataset, compared to the previous best floor-plan generation method. Furthermore, the resulting floor plans are multiple thousands of times smaller in memory size than the input point clouds, while still preserving the main building structures.",

keywords = "Floor-plan generation, indoor stereo SLAM, noisy point cloud filtering",

author = "Xin Liu and Egor Bondarau and \{de With\}, \{Peter H.N.\}",

year = "2023",

month = jun,

day = "7",

doi = "10.1117/12.2679358",

language = "English",

isbn = "9781510666184",

series = "Proceedings of SPIE",

publisher = "SPIE",

editor = "Wolfgang Osten and Dmitry Nikolaev and Jianhong Zhou",

booktitle = "Fifteenth International Conference on Machine Vision (ICMV 2022)",

address = "United States",

note = "15th International Conference on Machine Vision, ICMV 2022, ICMV 2022 ; Conference date: 18-11-2022 Through 20-11-2022",

url = "http://icmv.org/",

}

Floor-plan Generation from Noisy Point Clouds. / Liu, Xin; Bondarau, Egor ; de With, Peter H.N.
Fifteenth International Conference on Machine Vision (ICMV 2022). ed. / Wolfgang Osten; Dmitry Nikolaev; Jianhong Zhou. SPIE, 2023. 1270111 (Proceedings of SPIE; Vol. 12701).

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

TY - GEN

T1 - Floor-plan Generation from Noisy Point Clouds

AU - Liu, Xin

AU - Bondarau, Egor

AU - de With, Peter H.N.

N1 - Conference code: 15

PY - 2023/6/7

Y1 - 2023/6/7

N2 - This paper proposes a growing-based floor-plan generation method that creates the global layout of buildings from noisy point clouds obtained by a stereo camera. We introduce a PCA-based line-growing concept with a subsequent filtering step, which is able to robustly handle the high noise levels in input point clouds. Experimental results show that this method outperforms the state-of-the-art techniques in floor-plan generation. The average score for building layouts has increased from 0.38 to 0.66 on our test dataset, compared to the previous best floor-plan generation method. Furthermore, the resulting floor plans are multiple thousands of times smaller in memory size than the input point clouds, while still preserving the main building structures.

AB - This paper proposes a growing-based floor-plan generation method that creates the global layout of buildings from noisy point clouds obtained by a stereo camera. We introduce a PCA-based line-growing concept with a subsequent filtering step, which is able to robustly handle the high noise levels in input point clouds. Experimental results show that this method outperforms the state-of-the-art techniques in floor-plan generation. The average score for building layouts has increased from 0.38 to 0.66 on our test dataset, compared to the previous best floor-plan generation method. Furthermore, the resulting floor plans are multiple thousands of times smaller in memory size than the input point clouds, while still preserving the main building structures.

KW - Floor-plan generation

KW - indoor stereo SLAM

KW - noisy point cloud filtering

UR - https://www.scopus.com/pages/publications/85172928812

U2 - 10.1117/12.2679358

DO - 10.1117/12.2679358

M3 - Conference contribution

SN - 9781510666184

T3 - Proceedings of SPIE

BT - Fifteenth International Conference on Machine Vision (ICMV 2022)

A2 - Osten, Wolfgang

A2 - Nikolaev, Dmitry

A2 - Zhou, Jianhong

PB - SPIE

T2 - 15th International Conference on Machine Vision, ICMV 2022

Y2 - 18 November 2022 through 20 November 2022

ER -

Floor-plan Generation from Noisy Point Clouds

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this