UDMV: EG Workshop on Urban Data Modelling and Visualisation

Permanent URI for this community

https://diglib.eg.org/handle/10.2312/450

Browse

Browsing UDMV: EG Workshop on Urban Data Modelling and Visualisation by Subject "Boundary representations"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Indoor Scene Reconstruction using Primitive-driven Space Partitioning and Graph-cut
    (The Eurographics Association, 2013) Oesau, Sven; Lafarge, Florent; Alliez, Pierre; Vincent Tourre and Gonzalo Besuievsky
    We present a method for automatic reconstruction of permanent structures of indoor scenes, such as walls, floors and ceilings, from raw point clouds acquired by laser scanners. Our approach employs graph-cut to solve an inside/outside labeling of a space decomposition. To allow for an accurate reconstruction the space decomposition is aligned with permanent structures. A Hough Transform is applied for extracting the wall directions while allowing a flexible reconstruction of scenes. The graph-cut formulation takes into account data consistency through an inside/outside prediction for the cells of the space decomposition by stochastic ray casting, while favoring low geometric complexity of the model. Our experiments produces watertight reconstructed models of multi-level buildings and complex scenes.
  • Thumbnail Image
    Item
    Topological Reconstruction of Complex 3D Buildings and Automatic Extraction of Levels of Detail
    (The Eurographics Association, 2014) Diakité, Abdoulaye A.; Damiand, Guillaume; Maercke, Dirk Van; Gonzalo Besuievsky and Vincent Tourre
    This paper describes a new method allowing to retrieve the indoor and outdoor topology of a detailed 3D building model from its geometry and to extract different levels of detail (LoD) from the resulting topological description. No prior information about the initial model, except its geometric information is needed as input, and using the combinatorial maps data structure, the method recovers the topological information of the identified parts of the building. The topology is needed for most of the applications using 3D building models after the architects design it. While classical models available are mainly furnished in a Boundary Representation (B-Rep) format, we discuss how to recover the components that allow to distinguish the several parts of the building (defined as volumes) then the spatial relationships linking them.