Maarten Vergauwen

Maarten Vergauwen
Geomatics research group
Gebroeders De Smetstraat 1
9000 Gent
Belgium
room: 00.E003

tel:
+32 9 331 65 90
contact

Prof. dr. ir. Maarten Vergauwen is a professor and head of the Geomatics research group. His research interests include 3D reconstruction techniques, photogrammetry, surveying, GNSS, laser scanning and modeling. He received his PhD in the renowned VISICS research group of prof. Luc Van Gool where he did research on uncalibrated 3D reconstruction. He co-founded the mobile mapping spin-off company GeoAutomation, specialized in acquiring accurate 3D surveying information from mobile imagery. He is the co-author of over 20 papers in refereed journals and more than 70 papers in international conferences. He has participated in several EC funded projects, as well as multiple projects for ESA.

query=user:U0002399 year:[2014 TO 2025] &institution=lirias&from=1&step=20&sort=scdate
showing 1 to 20 of 86
Items per page 10 |20 |50
Sort newest first |author |title |popularity
  • thesis-dissertation
    De Geyter, Sam; 2025. Scan-to-BIM Automatisatie van as-built BIM productie d.m.v digitalisatie en machine learning.
    LIRIAS4241456
    description
    Currently there is a steep increase in demand for As-built Building Information Models (BIM) within the Architecture, Engineering and Construction (AEC) industry. These BIMs are mostly created with a Scan-to-BIM process. The conversion of the point cloud data captured with remote sensing techniques to a parametric BIM model is currently a manual, time-consuming and error-prone task. In this research, the modeling part of the process will be automated in consecutive steps. We will use the Lidar data as well as other information captured by modern day acquisition technologies. First, the building's structure will be automatically detected and modeled. These objects will then be used as prior knowledge to detect and model the secondary building elements such as windows and doors to the completion of an architectural as-built BIM. In a last step, object recognition algorithms will be used to detect appliances, such as power outlets, light armatures, fire extinguishers etc. in the point cloud and image data to the creation of as-built HVAC, MEP and FM BIM.

    Accepted
  • conference
    Vermandere, Jelle;Bassier, maarten;Vergauwen, maarten; 2025. Geometry and Texture Completion of Partially Scanned 3D Objects Through Material Segmentation. Proceedings of the 20th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - GRAPP; 2025; pp. 193 - 202
    LIRIAS4230127
    description
    This work aims to improve the geometry and texture completion of partially scanned 3D objects in indoor environments through the integration of a novel material prediction step. Completing segmented objects from these environments remains a significant challenge due to high occlusion levels and texture variance. State-of-the-art techniques in this field typically follow a two-step process, addressing geometry completion first, followed by texture completion. Although recent advancements have significantly improved geometry completion, texture completion continues to focus primarily on correcting minor defects or generating textures from scratch. This work highlights key limitations in existing completion techniques, such as the lack of material awareness, inadequate methods for fine detailing, and the limited availability of textured 3D object datasets. To address these gaps, a novel completion pipeline is proposed, enhancing both the geometry and texture completion processes. Experi mental results demonstrate that the proposed method produces clearer material boundaries, particularly on scanned objects, and generalizes effectively even with synthetic training data.

    Published online
  • journal-article
    Zhao, Wufan;Persello, Claudio;Lv, Xianwei;Stein, Alfred;Vergauwen, Maarten; 2024. Vectorizing planar roof structure from very high resolution remote sensing images using transformers. International Journal Of Digital Earth; 2024; Vol. 17; iss. 1; pp. 1 - 15
    LIRIAS4137439
    description
    Accurately predicting the geometric structure of a building's roof as a vectorized representation from a raster image is a challenging task in building reconstruction. In this paper, we propose an efficient and precise parsing method called Roof-Former, based on a vision Transformer. Our method involves three steps: (1) Image encoder and edge node initialization, (2) Image feature fusion with an enhanced segmentation refinement branch, and (3) Edge filtering and structural reasoning. Our method outperforms previous works on the vectorizing world building dataset and the Enschede dataset, with vertex and edge heat map F1-scores increasing from 87.1%, 76.2% to 89.1%, 78.1%, and from 69.7%, 68.8% to 71.2%, 69.5%, respectively. Furthermore, our method demonstrates superior performance compared to the current state-of-the-art based on qualitative evaluations, indicating its effectiveness in extracting global image information while maintaining the consistency and topological validity of the roof structure.
    Publisher: Taylor & Francis Group
    Published
  • thesis-dissertation
    Mattheuwsen, Lukas; 2024. Interpretation of mobile mapping data for the creation and update of spatial databases of the public domain.
    LIRIAS4157777
    description
    During a construction project on the public domain there are many different stakeholders who have to cooperate. To ensure this, a centralized reference database is needed. For this we have the GRB in Flanders, a spatial database which is used as reference for all project plans. Having such an important role, it is essential to keep the GRB up to date. The objective of this research project is to automatically create and update these spatial databases using mobile mapping data. This research project will investigate the use of cutting edge object detection and machine learning algorithms on imagery and LIDAR data from state of the art mobile mapping systems. This will result in an automated workflow for the creation of as-built maps used to create or update spatial databases and an error detection system to check existing databases. In addition, a later part will focus on the improvement of the data accuracy of low-end mobile mapping systems so they can be deployed for these mapping purposes.

    Published
  • presentation
    Aertsens, Tim;de Lima Hernandez, Roberto;Tnunay, Hilton;Vergauwen, Maarten;Monbaliu, Jaak;Toffoli, Alessandro; 2024. CORRELATION BETWEEN THE OBSERVED WHITE CAPPING AREA AND THE MEASURED SPECTRAL ENERGY DISSIPATION IN PHYSICAL BREAKING EXPERIMENTS.
    LIRIAS4222565
    description


    Published
  • thesis-dissertation
    Zalamea Patiño, Olga; 2023. Merging and Expanding existing Ontologies to cover the Built Cultural Heritage domain; case study Cuenca-Ecuador.
    LIRIAS4086112
    description
    Proper management of heritage cities is crucial for preserving and safeguarding their cultural value. However, achieving effective management requires the integration of data from various sources. This integration involves a vast amount of information, both thematic and spatial, which is currently managed separately. To address this issue, the use of Geographic Information Systems (GIS) and Building Information Modeling (BIM) technologies has gained popularity, as they provide tools for handling and analyzing diverse data at different scales. In the context of built cultural heritage management, traditional knowledge representation systems have limitations. These systems are typically centralized and rely on a shared understanding of concepts and terminology. However, in multidisciplinary domains , it is challenging to establish a common vocabulary and structure due to diverse perspectives and expertise. Thus, a flexible model is needed to harmonize the heterogeneous information associated with heritage cities and facilitate efficient information retrieval. One approach to address these challenges is the utilization of ontologies, which are conceptual models that define the relationships and properties of entities within a specific domain. Linked Open Data principles, which promote the publication and interlinking of structured data on the web, complement ontologies by facilitating the integration of information from various sources. In the cultural heritage domain, ontologies and linked open data have shown promise in managing intangible cultural heritage, movable heritage, and archaeological sites. However, their application to immovable heritage, such as historical buildings and monuments, requires further research to enable seamless integration with data from other domains. To bridge this gap, the research proposes the development of the BCH-ontology, a flexible ontological model specifically designed for managing built cultural heritage. The ontology aims to fulfill the multi-scale information requirements of heritage city management by reusing and extending existing ontologies. The chosen methodology for ontology development is the On-To-Knowledge approach, which consists of two processes: the Knowledge Meta process and the Knowledge process. In the Knowledge Meta process, terms related to preventive conservation are identified using a taxonomy. This taxonomy is then used to locate and incorporate relevant existing ontologies into the BCH-ontology. Specifically, three ontologies—the Geneva City Geographic Markup Language (Geneva-CityGML), Monument Damage ontology (MONDIS), and CIDOC Conceptual Reference Model (CIDOC-CRM)—are merged and expanded to create a comprehensive semantic framework for managing built cultural heritage. The Knowledge process focuses on validating and refining the BCH-ontology through practical use cases and stakeholder engagement. Data provided by stakeholders are translated into semantic representations and linked within the ontology, enabling basic queries to be executed. The ontology's usability and effectiveness are then explored through workshops with the stakeholders, facilitating further refinement and improvements. The application of the BCH-ontology is demonstrated in the context of the Historic Centre of Santa Ana de los Ríos de Cuenca in Ecuador, a UNESCO World Heritage site. Like many other heritage cities, Cuenca faces challenges in harmonizing and managing its cultural heritage information due to fragmented data storage and management practices. To overcome these challenges, Cuenca adopts a preventive conservation approach, which emphasizes regular risk and threat assessments to identify early signs of deterioration and minimize intervention. The development of the BCH-ontology provides a means to represent preventive conservation information effectively while accommodating alternative management approaches. By utilizing the ontology, stakeholders in Cuenca and other heritage cities can integrate diverse datasets and streamline their management processes. The ontology's potential applications extend beyond Cuenca, serving as a model for other heritage cities worldwide. The research highlights the broader significance of the BCH-ontology in the context of Ecuador's Open Data Law, which emphasizes the use of linked open data principles. Although the law is in effect, its practical implementation faces challenges due to a lack of clear examples. The BCH-ontology serves as a practical demonstration of implementing linked open data principles, providing guidance and promoting the effective utilization of open data in heritage city management. In conclusion, the proper management of heritage cities requires the integration of data and collaboration among multidisciplinary stakeholders. Ontologies, in conjunction with linked open data principles, offer a flexible and scalable solution for harmonizing heterogeneous heritage information. The development of the BCH-ontology addresses the specific needs of managing built cultural heritage, enabling effective information integration and retrieval. The application of the ontology in the Historic Centre of Cuenca demonstrates its practical value, while also contributing to the implementation of Ecuador's Open Data Law. By embracing the BCH-ontology and linked open data principles, heritage cities can enhance their management practices, preserve cultural heritage, and promote sustainable development.

    Published
  • conference
    Vincke, Stan;Mangnus, Dennis;Vergauwen, maarten; 2023. Image and BIM fusion database for visual construction progress monitoring. ACRS2023 Proceedings; 2023; pp. 1 - 10 Publisher: Asian Association on Remote Sensing (A.A.R.S.); Online [https://a-a-r-s.org/proceeding/ACRS2023/Spatial%20Information%20Sciences/ACRS2023024.pdf]
    LIRIAS4151263
    description
    The Architecture, Engineering and Construction (AEC) industry’s use of Building Information Modeling (BIM) has significantly increased over the past decade. Theoretically, this enables a fully digitalized construction approach where the executed constructional activities are crosschecked with the 4D as-design BIM model and its incorporated action schedule. However, in practice, this process is typically hampered by a lack of sufficient digital data representing the current state of the construction site. The proposed approach in this work aids in this challenge by providing a framework that not only fully automatically processes but also analyzes (daily) recorded construction images in terms of depicted BIM structures, hence fusing both formerly entirely separate worlds (real world vs. BIM environment). Moreover, an interactive platform is developed that allows users to digitally inspect the construction site in a comprehensible, yet efficient way. The backbone architecture uses an SQLite database that allows to efficiently store all recorded site information present in the imagery. Simultaneously, taking this approach also enables an opposite information flow back to the user in the shape of database queries. The developed framework is believed to increase the efficiency of constructional activity and progress monitoring, and, in general, yields a better oversight on the performed works. In a similar fashion as to what BIM meant in the digitalization process of the design stage of a construction undertaking, the presented framework could function as a comparable paradigm on the physical building and monitoring aspect.

    Published online
  • Sykora, Toon; 2023. The tomb of Djehutihotep within the Middle Kingdom elite cemetery of Dayr al-Barsha. Digital epigraphic documentation, analysis and interpretation..
    LIRIAS4086713
    description
    The rock-cut tombs that make up the elite cemetery at Dayr al-Barshā rank among the most important private monuments of the Egyptian Middle Kingdom (c. 2050-1750 BC). Especially the funerary complex of Djehutihotep is well-known for the quality of its decoration and the unusual subject-matter of some of the figurative scenes contained within. Unfortunately, the cemetery has suffered greatly from later phases of reuse and vandalism, which led to the collapse of many of the tombs and the degradation of their decoration. To overcome these methodological challenges, Puzzling Tombs, an interdisciplinary research project between engineers and Egyptologists of the KU Leuven was launched in October 2017. While the engineering track of this project explored the documentation and realignment of the fragmented architectural remains, our own study primarily focusses on the decoration in paint and relief. It presents a full digital epigraphic documentation analysis on the funerary chapel of Djehutihotep and a contextualised architectural study of the monument. An essential component of any epigraphic study is the investigation and integration of the archival record left by early visitors to the monument. For the tomb of Djehutihotep, we are fortunate to possess a rich collection of 17th to 19th century descriptions, drawings and squeezes that can aid in reconstructing parts of the structure and its history. For better preserved sections, the digital epigraphic method allows us to analyse and disseminate the content of and the interrelations between textual and iconographic elements of the original decoration in detail. By exposing these, the many Egyptological themes on which the scenes and texts in the tomb touch, can now be explored with much more fidelity to the original source than the highly antiquated publication record on the tomb previously allowed. When considered together, the decoration of Djehutihotep's tomb reveals a consistent and deliberate programme of self-representation by the tomb owner. Several architectural, iconographic and textual features in the tomb showcase a potent interplay between invention and tradition. This often extends to explicit archaic references not usually encountered to this degree or scale in private monuments of the Middle Kingdom. This well-considered programme of representation was not limited to the tomb owner, but also included secondary figures in the tomb, such as the chief artist, who used several visual ploys to emphasize his contribution in his own work. Part of the process that led to the creation of the tomb can be reconstructed by considering the tomb within its wider spatial setting. The architecture of the funerary complex was specifically adapted to the architectural history of the site and the geological properties of the rock in which it was carved. This demonstrates that the ancient surveyors were intimately aware of their physical environment and able to adjust accordingly. Although no unfinished sections of the tomb directly showcase earlier stages in its establishment, several preparatory traces visible underneath the final painted layer still allow us to discern some of the steps of the skilful artistic process that led to the decoration of the tomb. Other physical marks in and around the complex testify to the complex use-life of the tomb, from its initial establishment in the Middle Kingdom, through quarrying operations in the New Kingdom, Coptic reuse and looting and vandalism in the modern era. As well as a detailed investigation of a key-source in Middle Kingdom iconography, this study should be seen as a case study showcasing the potential of digital epigraphy as an analytical method. It can furthermore serve as a warning against any uncritical use of the often antiquated publications on primary sources, available to Egyptologists.

    Published
  • journal-article
    Symeonidis, Spyridon;Meditskos, Georgios;Vrochidis, Stefanos;Avgerinakis, Konstantinos;Derdaele, Jens;Vergauwen, Maarten;Bassier, Maarten;Moghnieh, Ayman;Fraguada, Luis;Vogler, Verena;Shekhawat, Yash;Wanner, Leo;Marimon, Montserrat;Valsamidou, Kalliopi;Koulali, Panagiota;Tellios, Anastasios;Wuyts, Jolan;Lopez, Eva; 2023. V4Design: Intelligent Analysis and Integration of Multimedia Content for Creative Industries. Ieee Systems Journal; 2023; Vol. 17; iss. 2; pp. 2570 - 2573
    LIRIAS3970469
    description

    Publisher: Institute of Electrical and Electronics Engineers
    Published
  • conference
    Vermandere, Jelle;Bassier, Maarten;Vergauwen, Maarten; 2023. MEASURE UP: A SERIOUS GAME FOR TOPOGRAPHIC EDUCATION. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences; 2023; Vol. X-5/W1-2023; pp. 69 - 74 Publisher: Copernicus Publications
    LIRIAS4085546
    description


    Published online
  • conference
    Bartczak, Erkki Tobias;Bassier, Maarten;Vergauwen, Maarten; 2023. CASE STUDY FOR UAS-ASSISTED BRIDGE INSPECTIONS. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences; 2023; Vol. XLVIII-2/W3-2023; pp. 33 - 39 Publisher: Copernicus Publications
    LIRIAS4083299
    description


    Published online
  • journal-article
    Cuypers, Suzanna;Nascetti, Andrea;Vergauwen, Maarten; 2023. Land Use and Land Cover Mapping with VHR and Multi-Temporal Sentinel-2 Imagery. Remote Sensing; 2023; Vol. 15; iss. 10
    LIRIAS4086490
    description

    Publisher: Multidisciplinary Digital Publishing Institute (MDPI)
    Published
  • Sykora, Toon;de Lima Hernandez, Roberto;De Meyer, Marleen;Vergauwen, Maarten;Willems, Harco; 2023. Puzzling Tombs: Virtual reconstruction of the Middle Kingdom elite necropolis at Dayr al-Barsha (Middle Egypt). Ancient Egypt and New Technology: The Present and Future of Computer Visualization, Virtual Reality and other Digital Humanities in Egyptology. Proceedings of a Conference held in Bloomington, Indiana, 29–30 March 2019; 2023; pp. 532 - 550 Publisher: Brill; Leiden
    LIRIAS3396123
    description


    Published
  • Cuypers, S;De Winter, H;Bassier, M;Vergauwen, M; 2023. PLANIMETRIC RAIL POSITIONING USING UAV PHOTOGRAMMETRY: TOWARDS AUTOMATED AND SAFE RAILWAY INFRASTRUCTURE MONITORING. GEOSPATIAL WEEK 2023, VOL. 48-1; 2023; pp. 589 - 596 Publisher: Copernicus Publications
    LIRIAS4155678
    description


    Published
  • conference
    De Winter, H;Bassier, Maarten;De Geyter, Sam;Vergauwen, Maarten; 2023. PHASED ACCURACY ANALYSIS IN ROAD CONSTRUCTION: USING BIM AND PHOTOGRAMMETRIC OUTPUT. GEOSPATIAL WEEK 2023; 2023; Vol. 48-1; pp. 1627 - 1634 Publisher: Copernicus Publications
    LIRIAS4155679
    description


    Published
  • Bartczak, ET;Bassier, M;Vergauwen, M; 2023. Proceedings in UAS-Assisted Bridge Inspections: RTK-Based Photogrammetric Reconstruction and Spatial Filtering. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences; 2023; Vol. XLVIII-1/W2-2023; pp. 1873 - 1880 Publisher: Copernicus Publications
    LIRIAS4155680
    description


    Published
  • conference
    De Geyter, Sam;Bassier, Maarten;De Winter, Heinder;Vergauwen, Maarten; 2023. OBJECT DETECTION AND LOCALISATION FOR BIM ENRICHMENT. GEOSPATIAL WEEK 2023; 2023; Vol. 48-1; pp. 155 - 162 Publisher: Copernicus Publications
    LIRIAS4155681
    description


    Published
  • conference
    Vermandere, Jelle;Bassier, Maarten;Vergauwen, Maarten; 2023. TEXTURE-BASED SEPARATION TO REFINE BUILDING MESHES. ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences; 2023; Vol. X-1/W1-2023; pp. 479 - 485 Publisher: Copernicus Publications
    LIRIAS4143867
    description


    Published online
  • journal-article
    Vincke, Stan;Vergauwen, Maarten; 2022. Vison based metric for quality control by comparing built reality to BIM. Automation In Construction; 2022; Vol. 144
    LIRIAS3928410
    description

    Publisher: Elsevier
    Published
  • thesis-dissertation
    de Lima Hernandez, Roberto; 2022. Puzzling Tombs. Digital reconstruction of the Egyptian Middle Kingdom governors' cemetery at Dayr al-Barshā.
    LIRIAS3768558
    description
    The rock tombs at Dayr al-Barsha rank among the most important monuments of the Egyptian Middle Kingdom (c. 2000-1700 BC). However, due to ancient quarrying and looting, this provincial elite cemetery is now in a dilapidated state. Thus making impossible to carry out a physical reconstruction on-site and consequently, a deep study of the tombs is prevented. Therefore, this research aims at digitally reconstructing the remains by using cutting-edge 3D acquisition techniques and novel registration algorithms. Which, not only will help to improve the puzzle-solving success but also allow for alteration as new elements surface, by means of an interactive 3D environment. With this applied science tool, fundamental research questions about the evolution of the cemetery and the use life and decoration of its tombs will be addressed. It will also serve as an aid for heritage preservation in an environment threatened due to the changing political climate in the Middle East.

    Published