Vectorizing planar roof structure from very high resolution remote sensing images using transformers
ABSTRACTAccurately predicting the geometric structure of a building's roof as a vectorized representation from a raster image is a challenging task in building reconstruction. In this paper, we propose an efficient and precise parsing method called Roof-Former, based on a vision Transformer. Ou...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Taylor & Francis Group
2024-12-01
|
Series: | International Journal of Digital Earth |
Subjects: | |
Online Access: | https://www.tandfonline.com/doi/10.1080/17538947.2023.2292637 |
_version_ | 1797387756984336384 |
---|---|
author | Wufan Zhao Claudio Persello Xianwei Lv Alfred Stein Maarten Vergauwen |
author_facet | Wufan Zhao Claudio Persello Xianwei Lv Alfred Stein Maarten Vergauwen |
author_sort | Wufan Zhao |
collection | DOAJ |
description | ABSTRACTAccurately predicting the geometric structure of a building's roof as a vectorized representation from a raster image is a challenging task in building reconstruction. In this paper, we propose an efficient and precise parsing method called Roof-Former, based on a vision Transformer. Our method involves three steps: (1) Image encoder and edge node initialization, (2) Image feature fusion with an enhanced segmentation refinement branch, and (3) Edge filtering and structural reasoning. Our method outperforms previous works on the vectorizing world building dataset and the Enschede dataset, with vertex and edge heat map F1-scores increasing from [Formula: see text], [Formula: see text] to [Formula: see text], [Formula: see text], and from [Formula: see text], [Formula: see text] to [Formula: see text], [Formula: see text], respectively. Furthermore, our method demonstrates superior performance compared to the current state-of-the-art based on qualitative evaluations, indicating its effectiveness in extracting global image information while maintaining the consistency and topological validity of the roof structure. |
first_indexed | 2024-03-08T22:30:42Z |
format | Article |
id | doaj.art-bf8ee8b44aa945b2b69b20fcd28bf7f0 |
institution | Directory Open Access Journal |
issn | 1753-8947 1753-8955 |
language | English |
last_indexed | 2024-03-08T22:30:42Z |
publishDate | 2024-12-01 |
publisher | Taylor & Francis Group |
record_format | Article |
series | International Journal of Digital Earth |
spelling | doaj.art-bf8ee8b44aa945b2b69b20fcd28bf7f02023-12-18T06:21:53ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552024-12-0117111510.1080/17538947.2023.2292637Vectorizing planar roof structure from very high resolution remote sensing images using transformersWufan Zhao0Claudio Persello1Xianwei Lv2Alfred Stein3Maarten Vergauwen4Geomatics Section, Department of Civil Engineering, Faculty of Engineering Technology, KU Leuven, Ghent, BelgiumDepartment of Earth observation science, Faculty of Geo-information Science and Earth Observation (ITC), University of Twente, Enschede, The NetherlandDepartment of Earth observation science, Faculty of Geo-information Science and Earth Observation (ITC), University of Twente, Enschede, The NetherlandDepartment of Earth observation science, Faculty of Geo-information Science and Earth Observation (ITC), University of Twente, Enschede, The NetherlandGeomatics Section, Department of Civil Engineering, Faculty of Engineering Technology, KU Leuven, Ghent, BelgiumABSTRACTAccurately predicting the geometric structure of a building's roof as a vectorized representation from a raster image is a challenging task in building reconstruction. In this paper, we propose an efficient and precise parsing method called Roof-Former, based on a vision Transformer. Our method involves three steps: (1) Image encoder and edge node initialization, (2) Image feature fusion with an enhanced segmentation refinement branch, and (3) Edge filtering and structural reasoning. Our method outperforms previous works on the vectorizing world building dataset and the Enschede dataset, with vertex and edge heat map F1-scores increasing from [Formula: see text], [Formula: see text] to [Formula: see text], [Formula: see text], and from [Formula: see text], [Formula: see text] to [Formula: see text], [Formula: see text], respectively. Furthermore, our method demonstrates superior performance compared to the current state-of-the-art based on qualitative evaluations, indicating its effectiveness in extracting global image information while maintaining the consistency and topological validity of the roof structure.https://www.tandfonline.com/doi/10.1080/17538947.2023.2292637Roof structure extractionvery high resolution remote sensingTransformergeometry reconstruction |
spellingShingle | Wufan Zhao Claudio Persello Xianwei Lv Alfred Stein Maarten Vergauwen Vectorizing planar roof structure from very high resolution remote sensing images using transformers International Journal of Digital Earth Roof structure extraction very high resolution remote sensing Transformer geometry reconstruction |
title | Vectorizing planar roof structure from very high resolution remote sensing images using transformers |
title_full | Vectorizing planar roof structure from very high resolution remote sensing images using transformers |
title_fullStr | Vectorizing planar roof structure from very high resolution remote sensing images using transformers |
title_full_unstemmed | Vectorizing planar roof structure from very high resolution remote sensing images using transformers |
title_short | Vectorizing planar roof structure from very high resolution remote sensing images using transformers |
title_sort | vectorizing planar roof structure from very high resolution remote sensing images using transformers |
topic | Roof structure extraction very high resolution remote sensing Transformer geometry reconstruction |
url | https://www.tandfonline.com/doi/10.1080/17538947.2023.2292637 |
work_keys_str_mv | AT wufanzhao vectorizingplanarroofstructurefromveryhighresolutionremotesensingimagesusingtransformers AT claudiopersello vectorizingplanarroofstructurefromveryhighresolutionremotesensingimagesusingtransformers AT xianweilv vectorizingplanarroofstructurefromveryhighresolutionremotesensingimagesusingtransformers AT alfredstein vectorizingplanarroofstructurefromveryhighresolutionremotesensingimagesusingtransformers AT maartenvergauwen vectorizingplanarroofstructurefromveryhighresolutionremotesensingimagesusingtransformers |