Abstract
Large-scale 3D scanning data based on point clouds enable accurate and fast recording of complex objects in the real world. The edges in a scanned point cloud usually describe the complex 3D structure of the target object and the surrounding scene. The recently proposed deep learning-based edge upsampling network can generate new points in the edge regions. When combined with the edge-highlighted transparent visualization method, this network can effectively improve the visibility of the edge regions in 3D-scanned point clouds. However, most previous upsampling experiments were performed on the sharp-edge regions despite that 3D-scanned objects usually contain both sharp and soft edge regions. In this paper, to demonstrate the performance of the upsampling network on soft-edge regions, we add more polygon models that contain soft edges by adjusting the models in the training set so that the network can learn more features of soft-edge regions. Additionally, we apply the upsampling network to real 3D-scanned point cloud data that contain numerous soft edges to verify that the edge upsampling network is equally effective at the upsampling task on soft-edge regions. The experimental results show that the visibility of the complex 3D-scanned objects can be effectively improved by increasing the point density in the soft-edge regions.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Kawakami, K., et al.: Opacity-based edge highlighting for transparent visualization of 3D scanned point clouds. ISPRS Ann. Photogramm. Remote. Sens. Spat. Inf. Sci. 5, 373–380 (2020)
Tanaka, S., et al.: See-through imaging of laser-scanned 3D cultural heritage objects based on stochastic rendering of large-scale point clouds. In: Proceedings of the ISPRS Annals of Photogrammetry, Remote Sensing & Spatial Information Sciences, Prague, Czech Republic, 12–19 July 2016, vol. III-3 (2016)
Uchida, T., et al.: Noise-robust transparent visualization of large-scale point clouds acquired by laser scanning. ISPRS J. Photogramm. Remote. Sens. 161, 124–134 (2020)
Yu, L., Li, X., Fu, C.-W., Cohen-Or, D., Heng, P.A.: PU-Net: point cloud upsampling network. In: Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA, 18–23 June 2018, pp. 2790–2799. Institute of Electrical and Electronics Engineers (IEEE), New York (2018)
Li, W., Hasegawa, K., Li, L., Tsukamoto, A., Tanaka, S.: Deep learning-based point upsampling for edge enhancement of 3D-scanned data and its application to transparent visualization. Remote Sens. 13, 2526 (2021)
West, K.F., Webb, B.N., Lersch, J.R., Pothier, S., Triscari, J.M., Iverson, A.E.: Context-driven automated target detection in 3D data. In: Proceedings of the Automatic Target Recognition XIV, Orlando, FL, USA, 21 September 2004 (2004)
Rusu, R.B.: Semantic 3D object maps for everyday manipulation in human living environments. KI Künstliche Intell. 24, 345–348 (2010)
Weinmann, M., Jutzi, B., Mallet, C.: Semantic 3D scene interpretation: a framework combining optimal neighborhood size selection with relevant features. ISPRS Ann. Photogramm. Remote. Sens. Spat. Inf. Sci. 2(3), 181–188 (2014)
Jutzi, B., Gross, H.: Nearest neighbour classification on laser point clouds to gain object structures from buildings. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 38, 4–7 (2009)
Corsini, M., Cignoni, P., Scopigno, R.: Efficient and flexible sampling with blue noise properties of triangular meshes. IEEE Trans. Vis. Comput. Graph. 18, 914–924 (2012)
Acknowledgments
The authors would like to thank the Tokushima Castle Museum for its cooperation in executing the 3D scanning. The images of the Borobudur temple are presented with the permission of the Borobudur Conservation Office and Research Center for Area Studies (P2W) of the Indonesian Institute of Sciences (LIPI). This work is partially supported by JSPS KAKENHI Grant Numbers 19KK0256 and 21H04903, and the Program for Asia-Japan Research Development (Ritsumeikan University, Japan).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Li, W. et al. (2022). Application of the Edge Upsampling Network to Soft-Edge Regions in a 3D-Scanned Point Cloud. In: Chang, BY., Choi, C. (eds) Methods and Applications for Modeling and Simulation of Complex Systems. AsiaSim 2021. Communications in Computer and Information Science, vol 1636. Springer, Singapore. https://doi.org/10.1007/978-981-19-6857-0_2
Download citation
DOI: https://doi.org/10.1007/978-981-19-6857-0_2
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-6856-3
Online ISBN: 978-981-19-6857-0
eBook Packages: Computer ScienceComputer Science (R0)