3D and Semantic Reconstruction of the Urban Environment Using Multi-Modal and Multi-Resolution Remote Sensing Data

Dear Colleagues,

Photogrammetry and remote sensing techniques are utilized to produce 3D models of urban scenes using satellite, aerial, and terrestrial data with varying levels of automation, accuracy, and replicability. These 3D models serve as the fundamental geometric structure for 3D geospatial environments and facilitate the integration of indoor data. The semantic information associated with 3D data enables spatiotemporal querying and analysis, making high-quality 3D semantic models valuable tools for environmental analysis and urban management.

The enormous amount of different modality remote sensing data available today requires powerful and efficient processing techniques to allow an ever-more accurate mapping of the urban environment at ever-higher spatial and temporal resolutions. Recently, advanced deep learning techniques have significantly improved the interpretation quality, efficiency, and automation levels in 3D scene understanding using remote sensing data. This Special Issue seeks to address the latest developments in remote sensing-based 3D urban scene reconstruction—from innovative methods and new benchmark datasets to relevant application examples.

This Special Issue welcomes contributions that showcase recent advancements in processing multimodal data for the geometric and semantic generation of 3D scenes, focusing on the urban environment. The entire pipeline, from data collection and processing to 3D object detection, modeling, and reconstruction and up to their representation, visualization, and application in urban environment analytics, is envisaged. We encourage the submission of open-source and open access methodologies and datasets.

The topics of this Special Issue include, but are not limited to:

• Weakly or self-supervised methods for extracting 3D semantic information of the urban environment;
• Multimodal approaches for combining different sensing technologies (e.g., multispectral, LiDAR, and SAR);
• Multiplatform (satellite, aerial, and terrestrial) and multiresolution data fusion approaches for 3D urban scene reconstruction;
• New benchmark datasets;
• Automatic 3D urban object identification and change detection methods from imagery, point clouds, and meshes;
• End-to-end approaches for the automatic generation of high-level semantic objects (e.g., LoD and BIM);
• Methods for efficient storage, processing, and visualization of 3D urban objects with a high level of detail and semantic attributes;
• Assessment of quality of 3D urban scene models, analysis of scalability, and replicability of proposed methods;
• Urban analytics based on multimodal remote sensing data and 3D building models.

Dr. Wufan Zhao
Dr. Andrea Nascetti
Prof. Dr. Maarten Vergauwen
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 100 words) can be sent to the Editorial Office for announcement on this website.

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.

• machine learning/deep learning
• data efficient learning (foundation model, transfer learning, self-supervised, semi-supervised, and weakly supervised learning)
• semantic/instance segmentation, object detection/tracking, and change detection
• SCAN-to-BIM
• data processing and feature extraction
• point cloud registration
• multisource data fusion
• application of 3D point clouds and remote sensing images
• spatial digital twins and smart cities