Smart City Gnosys
Smart city article details
| Title | 3D Urban Building Reconstruction Through Neural Rendering And Uav Imagery |
|---|---|
| ID_Doc | 186 |
| Authors | Cheolhwan K.; Youngmok K.; Wonjun C.; Eunkwan L.; Hong-Gyoo S. |
| Year | 2023 |
| Published | Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 41, 5 |
| DOI | http://dx.doi.org/10.7848/ksgpc.2023.41.5.407 |
| Abstract | The growing emphasis on smart cities and digital twins has heightened the significance of 3D geospatial information. Creating 3D models for urban buildings is a crucial area of interest but is severely labor-intensive and time-consuming. So, research has been conducted to reconstruct 3D urban structures automatically, employing various sensors. While there have been various studies on creating 3D models using UAV (Unmanned Aerial Vehicle) images, increasing the overlap between images to secure matching points and using an exponentially increased number of images for computation was mainly confirmed. Additionally, even when a sufficient number of input images were available, there were limitations in fully reconstructing all facets of a building, leading to instances where data from other sensor-based sources was needed. Therefore, in this study, we utilized only UAV images to create a point cloud using a deep learning-based neural rendering model and supplemented the occluded areas in conventional methodologies. We examined how processing time and Gaussian standard deviation change while generating building models by reducing the number of input images, comparing with ground truth derived from photogrammetry. Creating the 3D model with deep learning just took 25.251 minutes, a significant improvement from the 84.284 minutes required by the traditional method. Furthermore, even with a reduction of over three-quarters in the number of input images, the Gaussian standard deviation only showed a marginal difference of 0.203 meters compared to the entire image set. It underscores the effectiveness of our approach in supplementing occluded areas, surpassing the limitations observed in existing methods. © 2023 Korean Society of Surveying. All rights reserved. |
| Author Keywords | 3D Building Modeling; Deep Learning; Neural Radiance Field; Unmanned Aerial Vehicle |
Similar Articles
| Id | Similarity | Authors | Title | Published |
|---|---|---|---|---|
187  | 0.966 | Kim C.; Lee J.; Choi W.; Kwon Y.; Sohn H.-G. | 3D Urban Building Reconstruction Using Neural Rendering Technique | Computing in Civil Engineering 2023: Data, Sensing, and Analytics - Selected Papers from the ASCE International Conference on Computing in Civil Engineering 2023 (2024) |
| 748 | 0.924 | Chen X.; Cheng Y.; Han X.; Zhao B.; Tao W.; Ozdemir E.; Pan D. | A Compact Surface Reconstruction Method For Buildings Based On Convolutional Neural Network Fitting Implicit Representations | Journal of Computing in Civil Engineering, 39, 3 (2025) |
| 13602 | 0.9 | Akhavi Zadegan A.; Vivet D.; Hadachi A. | Challenges And Advancements In Image-Based 3D Reconstruction Of Large-Scale Urban Environments: A Review Of Deep Learning And Classical Methods | Frontiers in Computer Science, 7 (2025) |
| 13123 | 0.899 | Wang R.; Huang S.; Yang H. | Building3D: An Urban-Scale Dataset And Benchmarks For Learning Roof Structures From Point Clouds | Proceedings of the IEEE International Conference on Computer Vision (2023) |
| 59306 | 0.898 | Hu D.; Minner J. | Uavs And 3D City Modeling To Aid Urban Planning And Historic Preservation: A Systematic Review | Remote Sensing, 15, 23 (2023) |
| 47388 | 0.894 | Akbar A.; Liu C.; Wu H.; Jia S.; Xu Z. | Scene Information Guided Aerial Photogrammetric Mission Recomposition Towards Detailed Level Building Reconstruction | Advanced Engineering Informatics, 62 (2024) |
| 46426 | 0.891 | Zhang Z. | Review Of 3D Reconstruction Technology Of Uav Aerial Image | Journal of Physics: Conference Series, 1865, 4 (2021) |
| 23715 | 0.885 | Lee E.; Park S.; Jang H.; Choi W.; Sohn H.-G. | Enhancement Of Low-Cost Uav-Based Photogrammetric Point Cloud Using Mms Point Cloud And Oblique Images For 3D Urban Reconstruction | Measurement: Journal of the International Measurement Confederation, 226 (2024) |
| 15798 | 0.88 | Park J.K.; Heo J.H.; Lee K.W. | Construction Of Building 3D Model Using Drone Lidar Data; [드론 Lidar 데이터를 활용한 건물의 3D 모델 구축] | Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 43, 1 (2025) |
| 32001 | 0.878 | Wu C.; Yu X.; Ma C.; Zhong R.; Zhou X. | Integrating Geospatial Data And Street-View Imagery To Reconstruct Large-Scale 3D Urban Building Models | Transactions in GIS, 28, 5 (2024) |