Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting
True digital orthophoto maps (DOMs) are vital spatial data sources due to their high precision, detail, and accessibility. However, traditional generation methods using image differential correction with DEM/DSM often produce significant distortions from inaccurate surface data and missing building...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10930522/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849316088003690496 |
|---|---|
| author | Junxing Yang Zhenglong Cai Tianjiao Wang Tong Ye Haoran Gao He Huang |
| author_facet | Junxing Yang Zhenglong Cai Tianjiao Wang Tong Ye Haoran Gao He Huang |
| author_sort | Junxing Yang |
| collection | DOAJ |
| description | True digital orthophoto maps (DOMs) are vital spatial data sources due to their high precision, detail, and accessibility. However, traditional generation methods using image differential correction with DEM/DSM often produce significant distortions from inaccurate surface data and missing building information. Conventional geometric stitching and radiometric correction methods struggle to improve quality. While neural radiance field (NeRF)-based view synthesis offers progress, issues remain in training efficiency, rendering quality, and scene editability. To address these limitations, we propose Ortho-3DGS, a novel method for orthophoto generation from UAV imagery using 3D Gaussian Splatting (3DGS). Unlike NeRF, our approach models scenes via 3D Gaussian ellipsoids, optimized with depth supervision and gradient-based refinement for explicit and accurate reconstruction. We design a dedicated orthophoto rendering pipeline to generate high-quality, distortion-free DOMs efficiently. Experiments show Ortho-3DGS surpasses traditional tools (ContextCapture, Pix4Dmapper, Metashape) and NeRF-based methods (Ortho-NeRF, Ortho-NGP) in both radiometric and geometric performance. Ortho-3DGS offers commercial-grade accuracy while effectively mitigating distortions and artifacts, especially in complex environments. These results demonstrate its value for fast, precise orthophoto generation in diverse geospatial applications. |
| format | Article |
| id | doaj-art-cfb3d26d8ca34b14bfd88cd476d7012c |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-cfb3d26d8ca34b14bfd88cd476d7012c2025-08-20T03:51:58ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118109721099410.1109/JSTARS.2025.355210510930522Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian SplattingJunxing Yang0https://orcid.org/0000-0003-1893-3274Zhenglong Cai1https://orcid.org/0009-0002-3574-1143Tianjiao Wang2https://orcid.org/0009-0005-5729-4390Tong Ye3Haoran Gao4He Huang5https://orcid.org/0000-0003-3236-7539School of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing, ChinaSchool of Geomatics and Urban Spatial Informatics, Beijing University of Civil Engineering and Architecture, Beijing, ChinaTrue digital orthophoto maps (DOMs) are vital spatial data sources due to their high precision, detail, and accessibility. However, traditional generation methods using image differential correction with DEM/DSM often produce significant distortions from inaccurate surface data and missing building information. Conventional geometric stitching and radiometric correction methods struggle to improve quality. While neural radiance field (NeRF)-based view synthesis offers progress, issues remain in training efficiency, rendering quality, and scene editability. To address these limitations, we propose Ortho-3DGS, a novel method for orthophoto generation from UAV imagery using 3D Gaussian Splatting (3DGS). Unlike NeRF, our approach models scenes via 3D Gaussian ellipsoids, optimized with depth supervision and gradient-based refinement for explicit and accurate reconstruction. We design a dedicated orthophoto rendering pipeline to generate high-quality, distortion-free DOMs efficiently. Experiments show Ortho-3DGS surpasses traditional tools (ContextCapture, Pix4Dmapper, Metashape) and NeRF-based methods (Ortho-NeRF, Ortho-NGP) in both radiometric and geometric performance. Ortho-3DGS offers commercial-grade accuracy while effectively mitigating distortions and artifacts, especially in complex environments. These results demonstrate its value for fast, precise orthophoto generation in diverse geospatial applications.https://ieeexplore.ieee.org/document/10930522/3D Gaussian splatting (3DGS)depth supervisionthree-dimensional (3-D) reconstructiontrue orthodigital map (TDOM)unmanned aerial vehicles (UAVs) |
| spellingShingle | Junxing Yang Zhenglong Cai Tianjiao Wang Tong Ye Haoran Gao He Huang Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 3D Gaussian splatting (3DGS) depth supervision three-dimensional (3-D) reconstruction true orthodigital map (TDOM) unmanned aerial vehicles (UAVs) |
| title | Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting |
| title_full | Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting |
| title_fullStr | Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting |
| title_full_unstemmed | Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting |
| title_short | Ortho-3DGS: True Digital Orthophoto Generation From Unmanned Aerial Vehicle Imagery Using the Depth-Regulated 3D Gaussian Splatting |
| title_sort | ortho 3dgs true digital orthophoto generation from unmanned aerial vehicle imagery using the depth regulated 3d gaussian splatting |
| topic | 3D Gaussian splatting (3DGS) depth supervision three-dimensional (3-D) reconstruction true orthodigital map (TDOM) unmanned aerial vehicles (UAVs) |
| url | https://ieeexplore.ieee.org/document/10930522/ |
| work_keys_str_mv | AT junxingyang ortho3dgstruedigitalorthophotogenerationfromunmannedaerialvehicleimageryusingthedepthregulated3dgaussiansplatting AT zhenglongcai ortho3dgstruedigitalorthophotogenerationfromunmannedaerialvehicleimageryusingthedepthregulated3dgaussiansplatting AT tianjiaowang ortho3dgstruedigitalorthophotogenerationfromunmannedaerialvehicleimageryusingthedepthregulated3dgaussiansplatting AT tongye ortho3dgstruedigitalorthophotogenerationfromunmannedaerialvehicleimageryusingthedepthregulated3dgaussiansplatting AT haorangao ortho3dgstruedigitalorthophotogenerationfromunmannedaerialvehicleimageryusingthedepthregulated3dgaussiansplatting AT hehuang ortho3dgstruedigitalorthophotogenerationfromunmannedaerialvehicleimageryusingthedepthregulated3dgaussiansplatting |