Automatic 3d Building Model Generation Using Deep Learning Methods Based on Cityjson and 2d Floor Plans

Kippers, R. G.; Koeva, Mila; Keulen, Maurice van; Elberink, Sander Oude

doi:10.5194/isprs-archives-xlvi-4-w4-2021-49-2021

Cited by 10 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fully Convolutional Network-FCN) to detect initial building regions from LiDAR data and automatically reconstruct 3D prismatic building models from 3D LiDAR data. With the integration of 3D BAG CityJSON and floor plan images, Kippers et al (2021) proposed a new automatic DL-based method for constructing building models. An automatic 3D building reconstruction in LoD1 that consists of three main parts, DSM generation, Deep learningbased 2D building footprint generation, and 3D building reconstruction proposed by Yu et al (2020).…”

Section: Combination Of Dl-based and Conventional Methodsmentioning

confidence: 99%

Deep Learning for 3d Building Reconstruction: A Review

Buyukdemircioglu

Kocaman

Kada³

2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. 3D building reconstruction using Earth Observation (EO) data (aerial and satellite imagery, point clouds, etc.) is an important and active research topic in different fields, such as photogrammetry, remote sensing, computer vision and Geographic Information Systems (GIS). Nowadays 3D city models have become an essential part of 3D GIS environments and they can be used in many applications and analyses in urban areas. The conventional 3D building reconstruction methods depend heavily on the data quality and source; and manual efforts are still needed for generating the object models. Several tasks in photogrammetry and remote sensing have been revolutionized by using deep learning (DL) methods, such as image segmentation, classification, and 3D reconstruction. In this study, we provide a review on the state-of-the-art machine learning and in particular the DL methods for 3D building reconstruction for the purpose of city modelling using EO data. This is the first review with a focus on object model generation based on the DL methods and EO data. A brief overview of the recent building reconstruction studies with DL is also given. We have investigated the different DL architectures, such as convolutional neural networks (CNNs), generative adversarial networks (GANs), and the combinations of conventional approaches with DL in this paper and reported their advantages and disadvantages. An outlook on the future developments of 3D building modelling based on DL is also presented.

show abstract

Section: Combination Of Dl-based and Conventional Methodsmentioning

confidence: 99%

Deep Learning for 3d Building Reconstruction: A Review

Buyukdemircioglu

Kocaman

Kada³

2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…They used the Faster R-CNN framework to detect 6 classes on the same images and optical character recognition (OCR) to estimate room size. Kippers et al [15] developed an approach that combined semantic segmentation with U-Net and object detection with Faster R-CNN on the floor plan. They obtained the outline of the floor plan as a contour after the semantic segmentation task.…”

Section: Processing 2d Floor Plans Using Object Detection and Instanc...mentioning

confidence: 99%

Interoperability between Deep Neural Networks and 3D Architectural Modeling Software: Affordances of Detection and Segmentation

Wei,

Gupta,

Czerniawski

2023

Buildings

View full text Add to dashboard Cite

Building owners are working on converting their legacy documentation 2D floor plans into digital 3D representations, but the manual process is labor-intensive and time-consuming. In this paper, deep learning is leveraged to automate the process. This automation requires interoperability between artificial neural networks and prevailing 3D modeling software. The system processes 2D floor plans and outputs parameters of recognized walls, single doors, double doors, and columns. The parameters include the start point and end point of the wall and the center point of the door and column. These parameters are input into Revit 2022 through the Revit API 2022 after post-processing. The dimensional parameter integration affordances of object detection and instance segmentation are studied and compared using Faster R-CNN and Mask R-CNN models. Instance segmentation was found to require more time for data labeling but was more capable of informing the modeling of irregularly shaped objects. The mean Average Precision (mAP) of object detection and instance segmentation are 71.7% and 69.3%, respectively. Apart from single doors, the average precision for other categories falls within the range of 74% to 96%. The results provide software developers with guidance on choosing between object detection and instance segmentation strategies for processing legacy building documents. These types of systems are anticipated to be pivotal to the industry’s transition from 2D to 3D information modalities and advise practitioners to carefully choose suitable models and consider the recommendations provided in this study to mitigate potential failure cases.

show abstract

“…For the indoor environment of the building, techniques to automate the 3D modelling have also been presented over the years such as (Budroni and Boehm, 2010) where point clouds from scans are processed to create CAD models for 3D architecture or the method presented by (Boeters et al, 2015) where it automatically creates LoD2 models with indoor details, naming it LoD2+. A first approach integrating exterior and interior data sources as CityGML/JSON and 2D floor plan images for the reconstruction of 3D city models is proposed by (Kippers et al, 2021), which is based on deep learning methods. While promising results are obtained for less detailed and geometrically complex floor plans, the authors suggest including more representative training data to the machine learning model to improve their method.…”

Section: Related Workmentioning

confidence: 99%

Exploring Existing 3d Reconstruction Tools for the Generation of 3d City Models at Various Lod From a Single Data Source

Herrero

Ellul²,

Cavazzi³

2022

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. The use of three-dimensional (3D) city models has increased in a wide range of applications beyond visualisation. However, generation and maintenance of 3D data comes at a high cost, time, and workload. The purpose of the generalisation where coarser versions are obtained from a source data is of great interest for National Mapping and Cadastral Agencies (NMCA), which would benefit obtaining multiple 3D versions of an area from a single source. The main aim of the exploration presented in this paper is to study the potential of downsizing point clouds as an approach to generate 3D city models at multiple levels of detail from a single source and evaluate the steps required to ensure the output is fit for real world applications from an NMCA context. While interesting results are obtained when testing with sample data, no software managed to semi-automatically reconstruct 3D model for buildings of rather complicated geometry.

show abstract

Automatic 3d Building Model Generation Using Deep Learning Methods Based on Cityjson and 2d Floor Plans

Cited by 10 publications

References 16 publications

Deep Learning for 3d Building Reconstruction: A Review

Deep Learning for 3d Building Reconstruction: A Review

Interoperability between Deep Neural Networks and 3D Architectural Modeling Software: Affordances of Detection and Segmentation

Exploring Existing 3d Reconstruction Tools for the Generation of 3d City Models at Various Lod From a Single Data Source

Contact Info

Product

Resources

About