Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR)
This work proposes a method for creating accurate, watertight 3D building models from airborne laser scanning (ALS) point clouds by leveraging a modified Detection Transformer (DETR) architecture. We adapted the DETR architecture to directly predict building planes from point clouds, from which the...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Copernicus Publications
2025-07-01
|
| Series: | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| Online Access: | https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/773/2025/isprs-archives-XLVIII-G-2025-773-2025.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849421264877256704 |
|---|---|
| author | L. Kaufhold M. Kada |
| author_facet | L. Kaufhold M. Kada |
| author_sort | L. Kaufhold |
| collection | DOAJ |
| description | This work proposes a method for creating accurate, watertight 3D building models from airborne laser scanning (ALS) point clouds by leveraging a modified Detection Transformer (DETR) architecture. We adapted the DETR architecture to directly predict building planes from point clouds, from which the 3D model can be inferred using Boolean operations of half-spaces. We tested the model on the RoofN3D dataset and achieved a mean angle error of 1.7° for the building planes and a mean point-to-plane distance of 0.16m. Building facets can be detected even in the total absence of representative points, a common challenge in ALS data due to scanning direction and occlusions. By learning higher-level geometric principles, such as favouring 90-degree angles and symmetry, the model is able to adapt to various architectural styles without the need for explicit rules or pre-defined roof archetypes. |
| format | Article |
| id | doaj-art-e0cc3a0f6afb48f9a04c29dee0d13f05 |
| institution | Kabale University |
| issn | 1682-1750 2194-9034 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Copernicus Publications |
| record_format | Article |
| series | The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| spelling | doaj-art-e0cc3a0f6afb48f9a04c29dee0d13f052025-08-20T03:31:30ZengCopernicus PublicationsThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences1682-17502194-90342025-07-01XLVIII-G-202577378010.5194/isprs-archives-XLVIII-G-2025-773-2025Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR)L. Kaufhold0M. Kada1Institute of Geodesy and Geoinformation Science, Technische Universität Berlin, Berlin, GermanyInstitute of Geodesy and Geoinformation Science, Technische Universität Berlin, Berlin, GermanyThis work proposes a method for creating accurate, watertight 3D building models from airborne laser scanning (ALS) point clouds by leveraging a modified Detection Transformer (DETR) architecture. We adapted the DETR architecture to directly predict building planes from point clouds, from which the 3D model can be inferred using Boolean operations of half-spaces. We tested the model on the RoofN3D dataset and achieved a mean angle error of 1.7° for the building planes and a mean point-to-plane distance of 0.16m. Building facets can be detected even in the total absence of representative points, a common challenge in ALS data due to scanning direction and occlusions. By learning higher-level geometric principles, such as favouring 90-degree angles and symmetry, the model is able to adapt to various architectural styles without the need for explicit rules or pre-defined roof archetypes.https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/773/2025/isprs-archives-XLVIII-G-2025-773-2025.pdf |
| spellingShingle | L. Kaufhold M. Kada Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR) The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences |
| title | Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR) |
| title_full | Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR) |
| title_fullStr | Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR) |
| title_full_unstemmed | Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR) |
| title_short | Generating Watertight 3D Building Models from Airborne LiDAR Point Clouds using Detection Transformer (DETR) |
| title_sort | generating watertight 3d building models from airborne lidar point clouds using detection transformer detr |
| url | https://isprs-archives.copernicus.org/articles/XLVIII-G-2025/773/2025/isprs-archives-XLVIII-G-2025-773-2025.pdf |
| work_keys_str_mv | AT lkaufhold generatingwatertight3dbuildingmodelsfromairbornelidarpointcloudsusingdetectiontransformerdetr AT mkada generatingwatertight3dbuildingmodelsfromairbornelidarpointcloudsusingdetectiontransformerdetr |