ABSTRACT
3D surface reconstruction is the basis of reconstructing real objects in virtual world, it is widely used in digital twin, virtual reality and other fields. Complex physical structure of objects, unnatural spatial geometry characteristics, shadow effect caused by radar scanning and the uneven light will cause the edge distortion and point cloud void when the virtual environment is generated in indoor environment. These problems will cause the accuracy of 3D model seriously decreased in the reconstruction process above. In order to solve this problem, this paper proposed a Robust Moving Least Squares (RMLS) algorithm based on boundary separation. The point cloud is projected onto a two-dimensional plane after feature extraction of the initial point cloud data; the boundary of point cloud is extracted by grid partitioning method, the boundary region and point cloud body are separated; the RMLS algorithm is implemented on the boundary point cloud obtained by separation to increase the number of data points in the boundary region; besides, combine the point cloud with the triangulation surface reconstruction algorithm. The solution of problems reduced the edge distortion, made up the point cloud void and improved the reconstruction accuracy.
- Priyanka Mandikal, and Venkatesh Babu Radhakrishnan. 2019. Dense 3D Point Cloud Reconstruction Using a Deep Pyramid Network. 2019 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE, Waikoloa, 1052-1060. https://doi.org/10.1109/WACV.2019.00117Google ScholarCross Ref
- Maarten Bassier, and Maarten Vergauwen. 2020. Unsupervised reconstruction of Building Information Modeling wall objects from point cloud data. J. Automation in Construction 120, 103338 (September 2020), https://doi.org/10.1016/j.autcon.2020.103338Google ScholarCross Ref
- Marco Puliti, Giovanni Montaggioli, and Alessandro Sabato. 2021. Automated subsurface defects' detection using point cloud reconstruction from infrared images. J. Automation in Construction 129, 103829 (July 2021), https://doi.org/10.1016/j.autcon.2021.103829Google ScholarCross Ref
- K. L. Navaneet, Priyanka Mandikal, Mayank Agarwal, and R. Venkatesh Babu. 2019. CAPNet: Continuous Approximation Projection for 3D Point Cloud Reconstruction Using 2D Supervision. Proceedings of the AAAI Conference on Artificial Intelligence. AAAI, California, CA, 8819-8826. https://doi.org/10.1609/aaai.v33i01.33018819Google ScholarDigital Library
- Xu-Jia Qin, Zhong-Tian Hu, Hong-Bo Zheng, and Mei-Yu Zhang. 2019. Surface reconstruction from unorganized point clouds based on edge growing. J. Advances in Manufacturing 7, (September 2019), 343-352. https://doi.org/10.1007/s40436-019-00262-5Google ScholarCross Ref
- Junqiao Zhao, Qiaoyu Xu, Sisi Zlatanova, Liu Liu, Chen Ye, and Tiantian Feng. 2022. Weighted octree-based 3D indoor pathfinding for multiple locomotion types. J. International Journal of Applied Earth Observation and Geoinformation 112, 112900 (July 2022), https://doi.org/10.1016/j.jag.2022.102900Google ScholarCross Ref
- Jianli Du, Dong Chen, and Ruisheng Wang. 2019. A Novel Framework for 2.5-D Building Contouring From Large-Scale Residential Scenes. J. IEEE Transactions on Geoscience and Remote Sensing 57, 6 (June 2019), 4121-4145. https://doi.org/10.1109/TGRS.2019.2901539Google ScholarCross Ref
- Ahmad Mahphood, and Hossein Arefi, 2022. Grid-based building outline extraction from ready-made building points. J. Automation in Construction 139, 104321 (May 2022), https://doi.org/10.1016/j.autcon.2022.104321Google ScholarCross Ref
- Tey, W. Y., Che Sidik, N. A., Asako, Y., W Muhieldeen, M., and Afshar, O. 2021. Moving least squares method and its improvement: A concise review. J. Journal of Applied and Computational Mechanics 7, 2 (January 2021), 883-889. https://doi.org/10.22055/JACM.2021.35435.2652Google ScholarCross Ref
- Zhihai Liu, Hanbin Liu, Zhenguo Lu, and Qingliang Zeng. 2021. A Dynamic Fusion Pathfinding Algorithm Using Delaunay Triangulation and Improved A-Star for Mobile Robots. J. IEEE Access 9, (January 2021), 20602-20621. https://doi.org/10.1109/ACCESS.2021.3055231Google ScholarCross Ref
- S. R. Maurya, and G. M. Magar. 2018. Performance of Greedy Triangulation Algorithm on Reconstruction of Coastal Dune Surface. 2018 3rd International Conference for Convergence in Technology (I2CT). IEEE, Pune, 1-6. https://doi.org/10.1109/I2CT.2018.8529765Google ScholarCross Ref
- Ulug Bayazit. 2019. A greedy region growing algorithm for anisotropic stretch adaptive triangulation of geometry images. J. Graphical Models 106, 101045 (September 2019), https://doi.org/10.1016/j.gmod.2019.101045Google ScholarDigital Library
Index Terms
- Surface Reconstruction Based on Point Cloud Separation and Boundary RMLS
Recommendations
Tomographic surface reconstruction from point cloud
Inspired by computed tomography (CT), this paper presents a novel surface reconstruction algorithm, tomographic surface reconstruction, to reconstruct a surface mesh from a point cloud equipped with oriented normals. In the process of scanning a real ...
A Survey of Surface Reconstruction from Point Clouds
The area of surface reconstruction has seen substantial progress in the past two decades. The traditional problem addressed by surface reconstruction is to recover the digital representation of a physical shape that has been scanned, where the scanned ...
3D Surface Reconstruction from Point-and-Line Cloud
3DV '15: Proceedings of the 2015 International Conference on 3D VisionWe present a method for reconstructing 3D surface as triangular meshes from imagery. The surface reconstruction requires 3D point cloud for composing vertices of triangle meshes. A standard approach uses incremental structure from motion (SfM) to obtain ...
Comments