DEEP LEARNING FOR 3D BUILDING RECONSTRUCTION: A REVIEW

被引:18
作者
Buyukdemircioglu, Mehmet [1 ,2 ,4 ]
Kocaman, Sultan [2 ,3 ]
Kada, Martin [4 ]
机构
[1] Hacettepe Univ, Grad Sch Sci & Engn, Ankara, Turkey
[2] Hacettepe Univ, Dept Geomat Engn, Ankara, Turkey
[3] Swiss Fed Inst Technol, Inst Geodesy & Photogrammetry, CH-8093 Zurich, Switzerland
[4] Tech Univ Berlin, Inst Geodesy & Geoinformat Sci, Berlin, Germany
来源
XXIV ISPRS CONGRESS IMAGING TODAY, FORESEEING TOMORROW, COMMISSION II | 2022年 / 43-B2卷
关键词
Deep Learning; Machine Learning; 3D City Models; 3D Reconstruction; LoD; CITY MODELS; CLASSIFICATION;
D O I
10.5194/isprs-archives-XLIII-B2-2022-359-2022
中图分类号
P9 [自然地理学];
学科分类号
0705 ; 070501 ;
摘要
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.
引用
收藏
页码:359 / 366
页数:8
相关论文
共 71 条
[1]  
Agoub A., 2019, Int. Arch. Photogramm. Remote. Sens. Spat. Inf. Sci, VIV-4-W8, P3, DOI 10.5194/isprs-annals-IV-4-W8-3-2019
[2]  
Alidoost F., 2020, ISPRS ANN PHOTOGRAMM, V5, DOI [10.5194/isprsannals-V-2-2020-321-2020, DOI 10.5194/ISPRSANNALS-V-2-2020-321-2020]
[3]   2D Image-To-3D Model: Knowledge-Based 3D Building Reconstruction (3DBR) Using Single Aerial Images and Convolutional Neural Networks (CNNs) [J].
Alidoost, Fatemeh ;
Arefi, Hossein ;
Tombari, Federico .
REMOTE SENSING, 2019, 11 (19)
[4]  
[Anonymous], 2021, KAOLIN PYTORCH LIB A
[5]  
[Anonymous], 2012, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
[6]  
Beer L., 2019, GI Forum, V7, P119
[7]  
Biljecki F., 2019, ISPRS ANN PHOTOGRAMM, V4, P27, DOI DOI 10.5194/ISPRS-ANNALS-IV-4-W8-27-2019
[8]   Generating 3D city models without elevation data [J].
Biljecki, Filip ;
Ledoux, Hugo ;
Stoter, Jantien .
COMPUTERS ENVIRONMENT AND URBAN SYSTEMS, 2017, 64 :1-18
[9]   Applications of 3D City Models: State of the Art Review [J].
Biljecki, Filip ;
Stoter, Jantien ;
Ledoux, Hugo ;
Zlatanova, Sisi ;
Coeltekin, Arzu .
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION, 2015, 4 (04) :2842-2889
[10]  
Bittner K., 2018, Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci., V422, P103, DOI 10.5194/isprs-archives-XLII-2-103-2018