Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model

With the advancement of urbanization, building footprint data plays an important role in urban planning, 3D Real Scene and smart cities. Traditional manual contouring methods are time-consuming and laborious, while deep learning-based building extraction methods often require a large amount of label...

Full description

Saved in:
Bibliographic Details
Main Authors: J. Zhong, Y. Zhang, X. Liu, J. Zhang, L. Fei, W. Xia, B. Zhang, W. Fan, D. Yue
Format: Article
Language:English
Published: Copernicus Publications 2025-08-01
Series:The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Online Access:https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1777/2025/isprs-archives-XLVIII-G-2025-1777-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849397529760759808
author J. Zhong
Y. Zhang
Y. Zhang
X. Liu
X. Liu
J. Zhang
L. Fei
W. Xia
B. Zhang
W. Fan
D. Yue
author_facet J. Zhong
Y. Zhang
Y. Zhang
X. Liu
X. Liu
J. Zhang
L. Fei
W. Xia
B. Zhang
W. Fan
D. Yue
author_sort J. Zhong
collection DOAJ
description With the advancement of urbanization, building footprint data plays an important role in urban planning, 3D Real Scene and smart cities. Traditional manual contouring methods are time-consuming and laborious, while deep learning-based building extraction methods often require a large amount of labeled data and have limited generalization ability. In this paper, a zero-shot framework based on Segment Anything Model (SAM) is proposed for extracting and regularing building footprints from 3D mesh data. The method mainly consists of three steps: 1) Coarse Prompt Generation, irrelevant element’s masks such as ground and vegetation are eliminated by semi-global filtering and traditional classification method, and rough building mask is obtained as a boundary box prompt. 2) Fine mask generation: Using SAM's mask prompt capability, combined with logits map and grid elevation information with adaptive threshold to generate the fine mask prompt. Combine it with the updated bounding box to form hybrid prompt, and input SAM to generate a refined building mask. 3) Footprint regularization: Kinetic Partition, Markov random field, and Region Growth Algorithm are used to extract regularized building contours. Structural line segments from LSD guide the Kinetic Partitioning of the building. Markov random field matches building labels, while a region growth-based boundary reassignment refines the contours. The final regularized contour integrates the partitioned building zones. Our method achieved 78.31% AP50 on the Vaihingen dataset and obtained regular footprints that closely align with the true building contours on real Mesh data.
format Article
id doaj-art-c57b90d52d1c4e77822800fd0748d9ac
institution Kabale University
issn 1682-1750
2194-9034
language English
publishDate 2025-08-01
publisher Copernicus Publications
record_format Article
series The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
spelling doaj-art-c57b90d52d1c4e77822800fd0748d9ac2025-08-20T03:38:58ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342025-08-01XLVIII-G-20251777178310.5194/isprs-archives-XLVIII-G-2025-1777-2025Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh ModelJ. Zhong0Y. Zhang1Y. Zhang2X. Liu3X. Liu4J. Zhang5L. Fei6W. Xia7B. Zhang8W. Fan9D. Yue10School of Remote Sensing and Information Engineering, Wuhan University, Wuhan, 430079, Hubei, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, 430079, Hubei, ChinaTechnology Innovation Center for Collaborative Applications of Natural Resources Data in GBA, Ministry of Natural Resources, 510075, Guangzhou, Guangdong, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, 430079, Hubei, ChinaTechnology Innovation Center for Collaborative Applications of Natural Resources Data in GBA, Ministry of Natural Resources, 510075, Guangzhou, Guangdong, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, 100190, Beijing, ChinaChina Railway Siyuan Survey and Design Group Co., Ltd., 430063, Wuhan, Hubei, ChinaChina Railway Siyuan Survey and Design Group Co., Ltd., 430063, Wuhan, Hubei, ChinaChina Railway Siyuan Survey and Design Group Co., Ltd., 430063, Wuhan, Hubei, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, 430079, Hubei, ChinaSchool of Remote Sensing and Information Engineering, Wuhan University, Wuhan, 430079, Hubei, ChinaWith the advancement of urbanization, building footprint data plays an important role in urban planning, 3D Real Scene and smart cities. Traditional manual contouring methods are time-consuming and laborious, while deep learning-based building extraction methods often require a large amount of labeled data and have limited generalization ability. In this paper, a zero-shot framework based on Segment Anything Model (SAM) is proposed for extracting and regularing building footprints from 3D mesh data. The method mainly consists of three steps: 1) Coarse Prompt Generation, irrelevant element’s masks such as ground and vegetation are eliminated by semi-global filtering and traditional classification method, and rough building mask is obtained as a boundary box prompt. 2) Fine mask generation: Using SAM's mask prompt capability, combined with logits map and grid elevation information with adaptive threshold to generate the fine mask prompt. Combine it with the updated bounding box to form hybrid prompt, and input SAM to generate a refined building mask. 3) Footprint regularization: Kinetic Partition, Markov random field, and Region Growth Algorithm are used to extract regularized building contours. Structural line segments from LSD guide the Kinetic Partitioning of the building. Markov random field matches building labels, while a region growth-based boundary reassignment refines the contours. The final regularized contour integrates the partitioned building zones. Our method achieved 78.31% AP50 on the Vaihingen dataset and obtained regular footprints that closely align with the true building contours on real Mesh data.https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1777/2025/isprs-archives-XLVIII-G-2025-1777-2025.pdf
spellingShingle J. Zhong
Y. Zhang
Y. Zhang
X. Liu
X. Liu
J. Zhang
L. Fei
W. Xia
B. Zhang
W. Fan
D. Yue
Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model
The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
title Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model
title_full Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model
title_fullStr Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model
title_full_unstemmed Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model
title_short Zero-shot building footprint extraction and regularization based on Segment Anything model with Mesh Model
title_sort zero shot building footprint extraction and regularization based on segment anything model with mesh model
url https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/1777/2025/isprs-archives-XLVIII-G-2025-1777-2025.pdf
work_keys_str_mv AT jzhong zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT yzhang zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT yzhang zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT xliu zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT xliu zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT jzhang zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT lfei zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT wxia zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT bzhang zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT wfan zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel
AT dyue zeroshotbuildingfootprintextractionandregularizationbasedonsegmentanythingmodelwithmeshmodel