3-D Model Generation using Multi-View and Multi-Data
Keywords:
3D reconstruction, Data fusion, Digital elevation model (DEM), Multi-view stereo (MVS), Photogrammetry, Point cloud, Remote sensing, Structure-from-motion (SfM)Abstract
Generation of accurate three-dimensional (3D) models from satellite and geospatial data is an important task in remote sensing, photogrammetry, and geospatial analysis. Conventional single-view reconstruction techniques often suffer from incomplete terrain representation, limited depth information, and poor visualization quality. To address these challenges, this paper presents a multi-view and multi-data approach for generating realistic and detailed 3D terrain models. The proposed framework integrates Sentinel-1 SAR imagery, Sentinel-2 optical imagery, Digital Elevation Models (DEM), and Bhuvan geospatial datasets to provide complementary information related to terrain elevation, surface texture, and structural features. The key contribution of this work is the integration of multiple freely available datasets with photogrammetric techniques such as Structure-from-Motion (SfM) and Multi-View Stereo (MVS) to achieve improved terrain representation using a cost-effective workflow. The methodology includes data acquisition, preprocessing, feature extraction, feature matching, image registration, depth estimation, dense point cloud generation, surface reconstruction, and texture mapping. SNAP software is used for satellite image preprocessing and terrain correction, while QGIS is employed for geospatial analysis and 3D visualization. The framework was evaluated based on reconstruction completeness, elevation consistency, terrain representation quality, and visualization effectiveness. Experimental results demonstrated successful generation of DEM, DTM, DSM, river network extraction, and interactive 3D terrain models. The generated models are suitable for applications such as urban planning, environmental monitoring, disaster management, and geospatial analysis. Future work will focus on integrating deep learning techniques and higher-resolution datasets to further enhance reconstruction accuracy and automation.
References
J. Gao, J. Liu and S. Ji, “A general deep learning based framework for 3D reconstruction from multi-view stereo satellite images,” ISPRS Journal of Photogrammetry and Remote Sensing, vol. 195, pp. 446–461, Jan. 2023.
C. Stucker, B. Ke, Y. Yue, S. Huang, I. Armeni, and K. Schindler, “ImpliCity: City modeling from satellite images with deep implicit occupancy fields,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. V-2-2022, pp. 193–201, 2022.
F. Remondino and S. El-Hakim, “Image-based 3D modelling: A review,” The Photogrammetric Record, vol. 21, no. 115, pp. 269–291, Aug. 2006.
X. X. Zhu et al., “Deep learning in remote sensing: A comprehensive review and list of resources,” in IEEE Geoscience and Remote Sensing Magazine, vol. 5, no. 4, pp. 8–36, Dec. 2017.
E. Rupnik, M. Pierrot-Deseilligny and A Delorme, “3D reconstruction from multi-view VHR-satellite images in MicMac,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 139, pp. 201–211, May 2018.
R. Hartley and A. Zisserman, Multiple view geometry in computer vision, 2nd ed. Cambridge, U.K.: Cambridge University Press, 2004.
D. G. Lowe, “Distinctive image features from scale-invariant keypoints,” International Journal of Computer Vision, vol. 60, pp. 91–110, Nov. 2004.
H. Bay, T. Tuytelaars, and L. V. Gool, “SURF: Speeded up robust features,” Computer Vision - European Conference on Computer Vision (ECCV), 2006, pp. 404–417.
L. Zhang and E. Rupnik, “SparseSat-NeRF: Dense depth supervised neural radiance fields for sparse satellite images,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2023, pp. 895–902.
J. Wang, B. Sun and Y. Lu, “MVPNet: Multi-view point regression networks for 3D object reconstruction from a single image,” Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, no. 1, 2019, pp. 8949–8956.
Z. Li, B. Xu and J. Shan, “Geometric object based building reconstruction from satellite imagery derived point clouds,” The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLII-2/W13, pp. 73–78, 2019.
B. Kerbl, G. Kopanas, T. Leimkühler, and G. Drettakis, “3D Gaussian splatting for real-time radiance field rendering,” ACM Transactions on Graphics, vol. 42, no. 4, pp. 1–14, Jul. 2023.
L. Cheng, J. Gong, M. Li and Y. Liu, “3D building model reconstruction from multi-view aerial imagery and LiDAR data,” Photogrammetric Engineering & Remote Sensing, vol. 77, no. 2, pp. 125–139, Feb. 2011.
R. Qin, S. Song, and X. Huang, “3D data generation using low-cost cross-view images,” International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. XLIII-B2-2020, pp. 157–162, Aug. 2016.
S. Song and R. Qin, “Optimizing mesh reconstruction and texture mapping generated from a combined side-view and over-view imagery,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. V-2, pp. 403–409, 2020.
F. Rottensteiner, G. Sohn, J. Jung, M. Gerke, C. Baillard, S. Benitez, and U. Breitkopf, “ISPRS benchmark on urban 3D reconstruction and semantic modelling,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 1–3, pp. 293–298, 2012.
P. J. Besl and N. D. McKay, “A method for registration of 3-D shapes,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 239–256, Feb. 1992.
