loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Affiliation

Select a Subject

Institute for Pattern Recognition, Faculty of Electrical Engineering and Computer, Niederrhein University of Applied Sciences, Reinarzstr. 49, 47805 Krefeld and Germany
Context (please use complete words only)
Field
Sort by

4 Papers

Polyline Simplification with Predefined Edge Directions by Mixed Integer Linear Programs

Topics: Modeling and Algorithms; Scene and Object Modeling; Fundamental Methods and Algorithms; 3D Model Repair and scene reconstruction

Authors: Steffen Goebbels and Jochen Rethmann

Affiliation: Institute for Pattern Recognition, Faculty of Electrical Engineering and Computer Science, Niederrhein University of Applied Sciences, Reinarzstr. 49, 47805 Krefeld, Germany

Digital Object Identifier: 10.5220/0012263600003660

Proceedings of GRAPP 2024, ISBN: 978-989-758-679-8

Collapse Button Quick Abstract
Mixed integer linear programs are presented that simplify polylines such that edges follow only some predefined directions from a given set. Under this constraint, solutions are computed that are closest to the given vertices, or only close to the original data, but with a minimum number of edges, or with a minimum length. The algorithms are applied to 3D building modeling from point clouds. Boundaries of roof facets (roof polygons) are simplified by considering directions of roof plane gradients and intersection lines between roof planes.

Automatic Reconstruction of Roof Overhangs for 3D City Models

Topics: Geometric Computing; Texture Models, Analysis, and Synthesis; 3D Model Repair and scene reconstruction

Authors: Steffen Goebbels and Regina Pohle-Fröhlich

Affiliation: Institute for Pattern Recognition, Faculty of Electrical Engineering and Computer Science, Niederrhein University of Applied Sciences, Reinarzstr. 49, 47805 Krefeld, Germany

Digital Object Identifier: 10.5220/0011604200003417

Proceedings of GRAPP 2023/VISIGRAPP, ISBN: 978-989-758-634-7

Collapse Button Quick Abstract
Most current 3D city models, created automatically from cadastral and remote sensing data and represented in CityGML, do not include roof overhangs, although these overhangs are very characteristic for the appearance of buildings. This paper describes an algorithm that procedurally adds such overhangs. When a CityGML model is textured, the size of the overhangs is determined by recognizing overhangs in facade textures. In this case, the method only needs an already existing model in CityGML representation. Alternatively, if an additional point cloud (e.g., from airborne laser scanning) is available, this cloud can be utilized to calculate the overhang sizes. We compare the results of both methods.

RANSAC for Aligned Planes with Application to Roof Plane Detection in Point Clouds

Topics: Surface Modeling; Fundamental Methods and Algorithms; 3D Model Repair and scene reconstruction

Authors: Steffen Goebbels and Regina Pohle-Fröhlich

Affiliation: Institute for Pattern Recognition, Faculty of Electrical Engineering and Computer Science, Niederrhein University of Applied Sciences, Reinarzstr. 49, 47805 Krefeld, Germany

Digital Object Identifier: 10.5220/0008836301930200

Proceedings of GRAPP 2020/VISIGRAPP, ISBN: 978-989-758-402-2

Collapse Button Quick Abstract
Random Sample Consensus (RANSAC) is a standard algorithm to recognize planes in point clouds. It does not require additional context information. However, it might be applied in situations where results can be improved based on domain knowledge. Such a situation is 3D building reconstruction from airborne laser scanning data. The normals of many roof facets are orthogonal to footprint vectors. This specific property helps to estimate roof planes more precisely. The paper describes the adapted RANSAC algorithm. It can be also used in other applications in which planes are aligned to supporting vectors.

Roof Segmentation based on Deep Neural Networks

Topics: Segmentation and Grouping; Machine Learning Technologies for Vision; Deep Learning for Visual Understanding

Authors: Regina Pohle-Fröhlich 1 , Aaron Bohm 2 , Peer Ueberholz 2 , Maximilian Korb 1 and Steffen Goebbels 1

Affiliations: 1 Institute for Pattern Recognition, Faculty of Electrical Engineering and Computer, Niederrhein University of Applied Sciences, Reinarzstr. 49, 47805 Krefeld and Germany , 2 Institute for High Performance Computing, Faculty of Electrical Engineering and Computer Science, Niederrhein University of Applied Sciences, Reinarzstr. 49, 47805 Krefeld and Germany

Digital Object Identifier: 10.5220/0007343803260333

Proceedings of VISAPP 2019/VISIGRAPP, ISBN: 978-989-758-354-4

Collapse Button Quick Abstract
The given paper investigates deep neural networks (DNNs) for segmentation of roof regions in the context of 3D building reconstruction. Point clouds as well as derived depth and density images are used as input data. For 3D building model generation we follow a data driven approach, because it allows the reconstruction of roofs with more complex geometries than model driven methods. For this purpose, we need a preprocessing step that segments roof regions of buildings according to the orientation of their slopes. In this paper, we test three different DNNs and compare results with standard methods using thresholds either on gradients of 2D height maps or on point normals. For the application of a U-Net, we transform point clouds to structured 2D height maps, too. PointNet and PointNet++ directly accept unstructured point clouds as input data. Compared to classical gradient and normal based threshold methods, our experiments with U-Net and PointNet++ lead to better segmentation of roof structures.
Collapse Button Related Ontology Subjects/Areas/Topics (3)
Computer Vision, Visualization and Computer Graphics

Image and Video Analysis

Segmentation and Grouping

Topic(s):

Filter listed results, by

Publication date

From:
To:
Author
Locate Author in list below
Add Author(s) to filter:

- Science and Technology Publications, Lda.
RESOURCES

Proceedings

Papers

Authors

Ontology

CONTACTS

Science and Technology Publications, Lda
Avenida de S. Francisco Xavier, Lote 7 Cv. C,
2900-616 Setúbal, Portugal.

Phone: +351 265 520 185 (National fixed network call)
Fax: +351 265 520 186
Email: info@scitepress.org

EXTERNAL LINKS

PRIMORIS

INSTICC

SCITEVENTS

CROSSREF

PROCEEDINGS SUBMITTED FOR INDEXATION BY:

dblp

Ei Compendex

SCOPUS

Semantic Scholar

Google Scholar

Microsoft Academic