The Building Indoor & Outdoor Modelling (BIOM) project aims at automatic, simultaneous indoor and outdoor modelling of buildings from images and dense point clouds. We want to achieve a complete, geometrically accurate, semantically annotated but nonetheless lean 3D CAD representation of buildings and objects they contain in the form of a Building Information Models (BIM) that will help manage buildings in all their life cycle (renovation, simulation, deconstruction). We view indoor and outdoor building modelling as a joint process where both worlds fruitfully cooperate and benefit one another both in terms of semantics and geometry. The hope is that this holistic scene understanding and reconstruction approach will lead to more complete, correct, and geometrically accurate building models.
The first challenge will be to accommodate for heterogeneous data as full building modeling calls for data acquisition inside and outside the building but also from an aerial point of view to model roof. The BIOM project will also aim at exploiting the complementarity of image and LiDAR data. Another challenge is coping with incomplete data due to occlusions by furniture inside and urban and mobile objects outside. Last but not least, BIOM aims at modeling a large variety of architectural styles, different interior scene layouts, and a high amount of different objects that may be contained within the scene.
State-of-the-art approaches treat outdoor and indoor worlds separately: most indoor reconstruction approaches focus on detailed modelling of single rooms whereas only very few have dealt with 3D modelling of complete floors (under Manhattan world assumptions). To the best of our knowledge, no works have been proposed, yet, that model buildings outdoor and indoor simultaneously within one single comprehensive framework.
After separate analysis of indoor and outdoor data and their registration, we propose to formalize complex priors about the structure of buildings and included objects in a probabilistic fashion. A crucial unsolved problem in probabilistic modelling of dense, textured point clouds is how to take into account object-level context and topology of large, complex, cluttered 3D scenes. In this regard, our research shall investigate where the sweet spot lies between generative, procedural modelling and discriminative object labeling and generalisation or primitives based reconstruction. What is more, a good compromise of fast unsupervised and expressive, detailed supervised modelling and object recognition will be a major part of our research.
The BIOM project will investigate multi-scale application use cases and their specific modelling needs and develop operational methodologies for producing “Application/Domain Ready” BIM models in the standard CityGML/IFC formats. Use cases and proofs of concept covering different phases and aspects of the urban project will be experimented over two sites and industrialised through dedicated web services.
The development of new methodologies and services (such as simulations and optimizations) based on virtual clones of buildings can only address the very limited fragment of the stock for which a BIM model exists. The BIOM project will address this issue by providing a robust, comprehensive and reference methodology for data acquisition and reliable BIM modeling to address a broad range of applications: inventory and urban studies, life-cycle management, construction works, occupancy phase, consultancy and communication of urban information. We expect the BIOM project to have a significant potential of transmission of outcomes to industry products. IGN and CSTB will work with industrial partners, existing spin-offs, and public authorities on the increased valuation of the produced reference. An open call for initiatives stemming from such a reference will be released in direction of SME and start-ups in order to raise interest of new actors in the domain and generate innovating services.
Monsieur Bruno Vallet (Institut national de l'infromation géographique et forestière)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LIGM Laboratoire d'Informatique Gaspard-Monge
IGN Institut national de l'infromation géographique et forestière
Inria Sophia Antipolis-Méditerranée Centre de Recherche Inria Sophia Antipolis - Méditerranée
CSTB Centre Scientifique et Technique du Bâtiment
INSA-ICube Laboratoire des sciences de l'Ingénieur, de l'Informatique et de l'Imagerie
Help of the ANR 723,171 euros
Beginning and duration of the scientific project: - 48 Months