Publications of Florent Poux
Bookmark and Share    
Full Text
See detailApplications des relevés 3D au patrimoine, de l'acquisition au système d'information 3D
Hallot, Pierre ULiege; Billen, Roland ULiege; Poux, Florent ULiege et al

Conference (2016, September 23)

Les techniques permettant l’acquisition 3D du patrimoine bâti sont maintenant assez mures et démocratiques pour être utilisées à grande échelle. De plus en plus d’édifices font l’objet d’une sauvegarde ... [more ▼]

Les techniques permettant l’acquisition 3D du patrimoine bâti sont maintenant assez mures et démocratiques pour être utilisées à grande échelle. De plus en plus d’édifices font l’objet d’une sauvegarde numérique sous la forme de nuages de points colorés. Il reste néanmoins de nombreux défis à relever notamment pour : harmoniser les pratiques d’acquisition en terme de qualité et de précision, assurer l’interopérabilité entre les jeux de données ; exploiter pleinement l’information ponctuelle collectée ; et proposer une visualisation adéquate de ces grands jeux de données. Dans le cadre des développements réalisés à l’Unité de Géomatique de l’ULg, nous travaillons à la mise au point de systèmes d’informations permettant d’assurer un lien entre les informations sémantiques issues des applications archéologiques, architecturales, d’histoire de l’art avec les modèles 3D réalisés. La mise en place de ces systèmes d’informations passe par plusieurs étapes de recherche relatives à la sectorisation automatique, les structures de données, la définitions d’ontologies thématiques et finalement les règles sémiologiques de représentations dans un environnement virtuel 3D. Au cours de cette présentation, nous proposons de parcourir trois projets de recherche en cours au sein de l’Unité afin l’illustrer nos propos. Nous présenterons plus particulièrement le cas de la cathédrale Saint-Paul de Liège. [less ▲]

Detailed reference viewed: 170 (29 ULiège)
See detailCASTLE4D : The castle of Franchimont
Luczfalvy Jancsó, Andrea ULiege; Hoffsummer, Patrick ULiege; Billen, Roland ULiege et al

Poster (2016, August 16)

Detailed reference viewed: 74 (18 ULiège)
See detailLe patrimoine en 3D : de l’acquisition aux nuages de points intelligents
Billen, Roland ULiege; Poux, Florent ULiege; Hallot, Pierre ULiege et al

Scientific conference (2016, February 17)

Detailed reference viewed: 122 (31 ULiège)
Full Text
See detailCASTLE4D: An Archaeological Information System Based on 3D Point Clouds
Luczfalvy Jancsó, Andrea ULiege; Billen, Roland ULiege; Hoffsummer, Patrick ULiege et al

in Lerma, José Luis; Cabrelles, Miriam (Eds.) Proceedings of the Arqueológica 2.0 - 8th International Congress on Archaeology, Computer Graphics, Cultural Heritage and Innovation : Universitat Politècnica de València, Valence 5-7 septembre 2016 (2016)

Detailed reference viewed: 151 (32 ULiège)
Full Text
See detailSmart point cloud: Toward an intelligent documentation of our world
Poux, Florent ULiege; Billen, Roland ULiege

Conference (2015, December 17)

A point cloud is the output data of many sensors, such as Lidar and laser scanners, or the processing output from photogrammetry and SFM reconstructions. It is a 3D spatial representation of the ... [more ▼]

A point cloud is the output data of many sensors, such as Lidar and laser scanners, or the processing output from photogrammetry and SFM reconstructions. It is a 3D spatial representation of the environment through thousands, million or billions of points, where each single point as a position in space (X, Y, Z) It is safe to say that today, the technology is ready, and enable us to digitize very quickly the world as it stands. This RAW data is rapidly growing with the autonomous vehicles mapping our roads (LiDAR system equip it) and the robotics field using sensors to evolve in the environment. But how do we use this data? Professionnals have been struggling with point cloud, or delivering it to a client because it looks nice, but is there a real added value to visualization only? Can our clients use it at its full potential? [less ▲]

Detailed reference viewed: 148 (19 ULiège)
Full Text
See detailComparing photogrammetry and laser scanning: a 3D reconstruction of the Castle of Penne
Novel, Cyril; Keriven, Renaud; Poux, Florent ULiege et al

Conference (2015, November 25)

In the last decade, the progress in computing power – both CPU and GPU – has allowed the development of photogrammetry softwares. These softwares can turned a set of photos of an object – such as a ... [more ▼]

In the last decade, the progress in computing power – both CPU and GPU – has allowed the development of photogrammetry softwares. These softwares can turned a set of photos of an object – such as a building – into a 3D model. They compete with laser scanners which are usually used for scanning objects like buildings and landmarks. In this paper, we compare outcomes of the software Smart3DCapture[4] with laser scanner point clouds of various scenes in order to evaluate the accuracy of photogrammetry reconstruction [less ▲]

Detailed reference viewed: 162 (12 ULiège)
Full Text
See detailSegmentation semi-automatique pour le traitement de données 3D denses : application au patrimoine architectural
Poux, Florent ULiege; Hallot, Pierre ULiege; Jonlet, Benoît ULiege et al

in XYZ: la Revue de l'Association Française de Topographie (2014), 141(36), 69-75

Detailed reference viewed: 143 (34 ULiège)
See detailVers de nouvelles perspectives lasergrammétriques
Poux, Florent ULiege

Book published by EUE (2014)

3D Laser scanner are very fast acquisition systems, but the huge amount of data collected demands processing optimization and automation. Segmentation is an essential step in the production chain of point ... [more ▼]

3D Laser scanner are very fast acquisition systems, but the huge amount of data collected demands processing optimization and automation. Segmentation is an essential step in the production chain of point clouds processing. It defines the pertinence and accuracy of a future model reconstruction, as well as a high level analysis of any scene. Its goal is to extract from a large 3D dataset different groups of points that share a logical link to wisely consider them as one entity. The semi-automatic method developed extract shapes for an archeological study on medieval construction out of a point cloud acquired with a 3D Scanner laser, representing the cathedral Saint-Paul of Liège (Belgium). We obtain a segmented point cloud that represents elegantly the scene and make it a base for a worthy analysis and reconstruction. The development of a virtual reality application through a WebGL development to promote cultural heritage transcend what exist nowadays, and become a new way of transmitting history to future generations. [less ▲]

Detailed reference viewed: 53 (12 ULiège)
Full Text
See detailSemi-automatic and versatile point-cloud segmentation by region growing based on normal estimation
Poux, Florent ULiege; Billen, Roland ULiege

Poster (2013, September 06)

Segmentation is an essential step in the production chain of point-cloud processing. It defines the pertinence and accuracy of a future 3D model reconstruction, as well as a high level analysis of any ... [more ▼]

Segmentation is an essential step in the production chain of point-cloud processing. It defines the pertinence and accuracy of a future 3D model reconstruction, as well as a high level analysis of any scene. Its goal is to extract from a large 3D dataset different groups of points that share a logical link to wisely consider them as one entity. Manual segmentation of a 3D point cloud is extremely time consuming and imprecise for large datasets, and thus automated classification is the main focus of this paper. [less ▲]

Detailed reference viewed: 69 (6 ULiège)
Full Text
See detailRecording, Documentation and Cooperation for Cultural Heritage
Poux, Florent ULiege; Billen, Roland ULiege

Poster (2013, September 02)

Detailed reference viewed: 13 (3 ULiège)