Deep learning applications for point clouds in the construction industry

被引：5
作者：
Yue, Hongzhe [1 ]
Wang, Qian [1 ]
Zhao, Hongxiang [1 ]
Zeng, Ningshuang [1 ]
Tan, Yi [2 ]
机构：
[1] Southeast Univ, Sch Civil Engn, Nanjing 211189, Peoples R China
[2] Shenzhen Univ, Key Lab Resilient Infrastructures Coastal Cities, Minist Educ, Shenzhen 518060, Peoples R China
来源：
AUTOMATION IN CONSTRUCTION | 2024年 / 168卷
关键词：
Point cloud; Construction industry; Deep learning; Semantic segmentation; Data-related issues;
D O I：
10.1016/j.autcon.2024.105769
中图分类号：
TU [建筑科学];
学科分类号：
0813 ;
摘要：
Deep learning (DL) on point clouds holds significant potential in the construction industry, yet no comprehensive review has thoroughly summarized its applications and shortcomings. This paper presents a detailed review of the current applications of DL on point clouds in the construction industry, highlighting existing challenges, limitations, and future research directions. A two-stage literature search was conducted, resulting in the collection of 55 research papers published since 2020. The review provides an overview of DL algorithms and examines the datasets used for DL on point clouds, including both real-world and synthetic datasets. Furthermore, it summarizes the various applications of DL on point clouds within the construction sector. Following this analysis, the paper discusses current deficiencies and potential improvements in model performance and datarelated issues. Finally, several recommendations are provided to advance the development of DL-based point cloud applications in the construction industry.
引用
收藏
页数：17
相关论文
共 99 条
  • [1] Baduge S.K., Thilakarathna S., Perera J.S., Arashpour M., Sharafi P., Teodosio B., Shringi A., Mendis P., Artificial intelligence and smart vision for building and construction 4.0: machine and deep learning methods and applications, Automation in Construction, 141, (2022)
  • [2] Perrier N., Bled A., Bourgault M., Cousin N., Danjou C., Pellerin R., Roland T., Construction 4.0: a survey of research trends, Journal of Information Technology in Construction, 25, pp. 416-437, (2020)
  • [3] Wang Q., Kim M.K., Applications of 3D point cloud data in the construction industry: a fifteen-year review from 2004 to 2018, Advanced Engineering Informatics, 39, pp. 306-319, (2019)
  • [4] Zhao K., Larsen A.J.A.P.A., Challenges and Opportunities in 3D Content Generation, (2024)
  • [5] Oechsle M., Peng S.Y., Geiger A., Ieee, UNISURF: Unifying neural implicit surfaces and radiance fields for multi-view reconstruction, 18th IEEE/CVF International Conference on Computer Vision (ICCV)Electr Network, pp. 5569-5579, (2021)
  • [6] Takikawa T., Litalien J., Yin K.X., Kreis K., Loop C., Nowrouzezahrai D., Jacobson A., McGuire M., Fidler S., S.O.C. IEEE Comp, Neural Geometric Level of Detail: Real-Time Rendering with Implicit 3D Shapes, IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)Electr Network, pp. 11353-11362, (2021)
  • [7] Yin C., Wang B., Gan V.J., Wang M., Cheng J.C., Automated semantic segmentation of industrial point clouds using ResPointNet++, Automation in Construction, 130, (2021)
  • [8] Wang B.Y., Wang Q., Cheng J.C.P., Song C.H., Yin C., Vision-assisted BIM reconstruction from 3D LiDAR point clouds for MEP scenes, Automation in Construction, 133, (2022)
  • [9] Zhou Y.X., Ji A.K., Zhang L.M., Xue X.L., Attention-enhanced sampling point cloud network (ASPCNet) for efficient 3D tunnel semantic segmentation, Automation in Construction, 146, (2023)
  • [10] Fang X., Li H., Wu H.T., Fan L., Kong T., Wu Y., A fast end-to-end method for automatic interior progress evaluation using panoramic images, Engineering Applications of Artificial Intelligence, 126, (2023)
12345678910