PDF (10.2 MB)
Collect
Submit Manuscript
Research Article | Open Access

A new method for reconstructing building model using machine learning

Shengjie WuaHaibo YeaAntao LibHuawei TucShenxin XubDong Lianga()
College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
Shanghai Institute of Satellite Engineering, Shanghai 201109, China
Department of Computer Science and Information Technology, La Trobe University, Melbourne 3086, Australia
Show Author Information

Abstract

Three-dimensional (3D) model reconstruction is used in an increasing number of fields related to construction, such as urban planning, mobile communication planning, and solar power assessment. Existing 3D reconstruction models mostly focus on precise measurements, such as laser scanning and ultrasonic mapping. Although these methods can achieve very precise results, they require specific equipment, which is typically expensive. The essence of 3D reconstruction is to infer the overall view of a building through pictures from previously taken perspectives, thereby obtaining pictures from unfamiliar perspectives. In this study, the rendering method is adopted as the starting point, and architectural features are learned by training a neural network to provide the necessary information for rendering. Unlike the more popular projection-based raster rendering method, this study uses a point-based volume rendering method and light sampling to detect architectural features. This rendering method requires the color and density of specific sampling points. Therefore, this study attempts to train a neural network to fit a five-dimensional (5D) function. The input to this function is a 5D vector, including the position (x, y, z) and viewing direction (θ, φ), and the output is the color and density of this point when viewed from this direction. This study adopts the positional encoding method, which reduces the scale of the network and increases both the training and rendering speeds. Our method can train a usable network in dozens of seconds and render a building at 30–60 frames per second.

References

[1]

A. Agoub, V. Schmidt, M. Kada. Generating 3D city models based on the semantic segmentation of lidar data using convolutional neural networks. ISPRS Ann Photogramm Remote Sens Spatial Inf Sci, 2019, IV-4/W8: 3–10.

[2]

C. Kropp, C. Koch, M. König. Interior construction state recognition with 4D BIM registered image sequences. Autom Constr, 2018, 86: 11–32.

[3]

T. Czerniawski, F. Leite. Automated digital modeling of existing buildings: A review of visual object recognition methods. Autom Constr, 2020, 113: 103131.

[4]

C. Brenner. Building reconstruction from images and laser scanning. Int J Appl Earth Obs Geoinf, 2005, 6: 187–198.

[5]

C. Rocchini, P. Cignoni, C. Montani, et al. A low cost 3D scanner based on structured light. Comput Graph Forum, 2001, 20: 299–308.

[6]

C. W. Yao. An ultrasonic method for 3D reconstruction of surface topography. J Phys Commun, 2018, 2: 055034.

[7]

S. Hosseinian, H. Arefi. 3D reconstruction from multi-view medical X-ray images-review and evaluation of existing methods. Int Arch Photogramm Remote Sens Spatial Inf Sci, 2015, XL-1/W5: 319–326.

[8]
C. Y. Jiang, A. Sud, A. Makadia, et al. Local implicit grid representations for 3D scenes. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, USA, 2020: pp 6000–6010.
[9]
J. J. Park, P. Florence, J. Straub, et al. DeepSDF: Learning continuous signed distance functions for shape representation. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: pp 165–174.
[10]
K. Genova, F. Cole, A. Sud, et al. Local deep implicit functions for 3D shape. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, USA, 2020: pp 4857–4866.
[11]
L. Mescheder, M. Oechsle, M. Niemeyer, et al. Occupancy networks: Learning 3D reconstruction in function space. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, USA, 2019: pp 4460–4470.
[12]
Y. Yang, D. C. Zhan, Y. Fan, et al. Deep learning for fixed model reuse. In: Proceedings of AAAI Conference on Artificial Intelligence, San Francisco, USA, 2017: 2831–2837.
[13]
Y. Yang, D. W. Zhou, D. C. Zhan, et al. Adaptive deep models for incremental learning: Considering capacity scalability and sustainability. In: Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Anchorage, USA, 2019: pp 74–82.
[14]
A. X. Chang, T. Funkhouser, L. Guibas, et al. ShapeNet: An information-rich 3D model repository. 2015, arXiv: 1512.03012. arXiv.org e-print archive. https://arxiv.org/abs/1512.03012 (accessed 2024-6-24
[15]
D. M. Harris, S. L. Harris. State encodings. In: Digital Design and Computer Architecture. 2nd ed. D. M. Harris, S. L. Harris, Eds. Boston, USA: Morgan Kaufmann, 2013: pp 129–131.
[16]

O. Kurban, N. Calik, T. Yildirm. Human and action recognition using adaptive energy images. Pattern Recogn, 2022, 127: 108621.

[17]
R. Chabra, J. E. Lenssen, E. Ilg, et al. Deep local shapes: Learning local SDF priors for detailed 3D reconstruction. In: Proceedings of the 16th European Conference on Computer Vision. Glasgow, UK, 2020: 608–625.
[18]
T. Takikawa, J. Litalien, K. X. Yin, et al. Neural geometric level of detail: Real-time rendering with implicit 3D shapes. In: Proceedings of IEEE/CVF Conference on Computer Vision and Pattern Recognition, Nashville, USA, 2021: pp 11358–11367.
[19]
L. Li, D. Liang, Y. H. Gao, et al. ALL-E: Aesthetics-guided low-light image enhancement. In: Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence, Macao, China, 2023: 1062–1070.
[20]
Y. Y. Zhou, D. Liang, S. C. Chen, et al. Improving lens flare removal with general-purpose pipeline and multiple light sources recovery. In: Proceedings of IEEE/CVF International Conference on Computer Vision, Paris, France, 2023: pp 12969–12979.
[21]

M. Levoy. Efficient ray tracing of volume data. ACM Trans Graph, 1990, 9: 245–261.

Journal of Intelligent Construction
Article number: 9180041
Cite this article:
Wu S, Ye H, Li A, et al. A new method for reconstructing building model using machine learning. Journal of Intelligent Construction, 2025, 3(1): 9180041. https://doi.org/10.26599/JIC.2025.9180041
Metrics & Citations  
Article History
Copyright
Rights and Permissions
Return