AI-Enhanced Photogrammetry and Remote Sensing for Image-Based 3D Reconstruction

Dear Colleagues,

Recent advances in artificial intelligence (AI) are profoundly reshaping photogrammetry and remote sensing, particularly in the context of image-based 3D reconstruction. Machine learning and data-driven approaches are increasingly integrated into photogrammetric processes such as feature extraction, image matching, dense reconstruction, and semantic interpretation, while maintaining geospatial consistency and metric reliability.

In parallel, the rapid growth of imaging from aerial, terrestrial, UAV, and mobile mapping system has strengthened the role of remote sensing in high-resolution 3D modelling. The integration of multi-sensor datasets within AI-enhanced photogrammetric frameworks enables more automated, scalable, and information-rich reconstruction processes across a wide range of spatial scales. Emerging paradigms such as neural implicit representations (e.g., Neural Radiance Fields (NeRF) and 3D Gaussian Splatting) are opening new perspectives for high-fidelity and computationally efficient 3D modelling.

These developments benefit applications ranging from environmental monitoring to civil and structural engineering, cultural heritage documentation, structure and infrastructure management, disaster risk assessment and urban analytics. The convergence of AI, imagery, and photogrammetric principles represents a major frontier for producing reliable and scalable image-based 3D reconstruction within geospatial sciences.

This Special Issue aims to collect high-quality research on the integration of Artificial Intelligence within rigorous photogrammetric and remote sensing frameworks. The objective is to strengthen image-based 3D reconstruction methodologies by combining AI-driven approaches with established principles of sensor modelling, geometric accuracy, spatial referencing, and quantitative validation.

The issue welcomes contributions addressing the full reconstruction pipeline—from image acquisition using UAV, airborne, satellite, terrestrial, and mobile mapping platforms to camera calibration, orientation and bundle adjustment, dense image matching, surface reconstruction, and semantic enrichment. AI-supported solutions should demonstrate improvements in automation, robustness, scalability, and uncertainty assessment while maintaining the metric reliability required in geospatial and documentation contexts.

The Special Issue especially encourages research that leverages multi-platform and multi-sensor remote sensing data, including airborne and satellite imagery, UAV-based acquisitions, mobile mapping systems, and LiDAR integration, to address data fusion, multi-temporal analysis, and large-scale 3D mapping of natural and built environments. Works that bridge methodological innovation with practical applications in heritage conservation, infrastructure management, and landscape analysis are particularly aligned with the scope.

Relevant topics include the following: AI-assisted GIS or HBIM generation; semantic segmentation of heritage assets; multi-temporal analysis for conservation monitoring; integration of AI within UAV and terrestrial photogrammetry; applications of NeRF; Gaussian Splatting; or hybrid neural–geometric approaches for high-fidelity, semantically rich, digital twins.

Specific attention will also be given to applications in cultural heritage documentation and monitoring, including studies focusing on the 3D reconstruction and multi-temporal assessment of archaeological sites, historical urban fabrics, and cultural landscapes using airborne, UAV, terrestrial or remote sensing data. AI-assisted HBIM generation, semantic enrichment of georeferenced heritage models, and integration of satellite or aerial data for landscape-scale heritage monitoring are strongly aligned with the scope of this Special Issue, which aims to strengthen the connection between AI-enhanced photogrammetry and remote sensing sciences, promoting reproducible, scalable, and geospatially consistent approaches for 3D data observation and heritage-informed spatial analysis.

• AI-enhanced photogrammetric processing for airborne, satellite, and UAV imagery;
• Sensor modelling, calibration, and orientation in AI-supported photogrammetry;
• Neural-hybrid approaches and novel view synthesis techniques for georeferenced 3D reconstruction within photogrammetric frameworks (e.g., 3D Gaussian Splatting, diffusion models, Neural Radiance Fields, Image-based rendering...);
• Machine learning methods for dense image matching and Multi-View Stereo (MVS) in remote sensing;
• High-resolution modelling for environmental monitoring, heritage documentation, and land surface processes;
• AI-based automation of the photogrammetric pipeline (calibration, orientation, dense and semantic reconstruction);
• Multi-sensor data fusion for geospatial modelling;
• Digital Surface Model (DSM) and Digital Terrain Model (DTM) generation using AI-supported techniques;
• Accuracy assessment, benchmarking, and uncertainty analysis of AI-based photogrammetric products;
• AI-assisted processing of UAV and mobile mapping imagery for the digital documentation of complex sites;
• Visualisation, immersive environments, and mixed-reality applications for 3D data exploration

Scalable computing frameworks for large-area photogrammetric mapping.

Dr. Valeria Croce
Dr. Francesca Matrone
Dr. Arnadi Murtiyoso
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• AI-enhanced photogrammetry
• image-based 3D reconstruction
• structure-from-motion (SfM) and multi-view stereo (MVS)
• 3D Gaussian Splatting and neural implicit representations
• LiDAR–imagery fusion
• UAV and airborne mapping
• digital heritage (GeoBIM, HBIM, digital twins, digital shadows)
• digital surface modelling (DSM/DTM)
• geospatial data fusion