Raoul Wessel

Automatic reconstruction of parametric building models from indoor point clouds

We present an automatic approach for the reconstruction of parametric 3D building models from indoor point clouds. While recently developed methods in this domain focus on mere local surface reconstructions which enable e.g. efficient visualization, our approach aims for a volumetric, parametric building model that additionally incorporates contextual information such as global wall connectivity. In contrast to pure surface reconstructions, our representation thereby allows more comprehensive use: first, it enables efficient high-level editing operations in terms of e.g. wall removal or room reshaping which always result in a topologically consistent representation. Second, it enables easy taking of measurements like e.g. determining wall thickness or room areas. These properties render our reconstruction method especially beneficial to architects or engineers for planning renovation or retrofitting. Following the idea of previous approaches, the reconstruction task is cast as a labeling problem which is solved by an energy minimization. This global optimization approach allows for the reconstruction of wall elements shared between rooms while simultaneously maintaining plausible connectivity between all wall elements. An automatic prior segmentation of the point clouds into rooms and outside area filters large-scale outliers and yields priors for the definition of labeling costs for the energy minimization. The reconstructed model is further enriched by detected doors and windows. We demonstrate the applicability and reconstruction power of our new approach on a variety of complex real-world datasets requiring little or no parameter adjustment. Graphical abstractDisplay Omitted HighlightsReconstruction of high-level, parametric building models from indoor scans.Goes beyond mere surface reconstruction of previous reconstruction methods.Plausible configuration of walls determined by means of global optimization.Enables intuitive editing on level of building elements, e.g. moving whole walls.

A 3D shape benchmark for retrieval and automatic classification of architectural data

When drafting new buildings, architects make intensive use of existing 3D models including building elements, furnishing, and environment elements. These models are either directly included into the draft or serve as a source for inspiration. To allow efficient reuse of existing 3D models, shape retrieval methods considering the specific requirements of architects must be developed. Unfortunately, common 3D shape benchmarks which are used to evaluate the performance of retrieval algorithms are not well suited for architectural data. First, they incorporate models which are not related to this domain, and second and even more important, the provided classification schemes usually do not match an architects intuition regarding their notion of design and function. To overcome these drawbacks, we present a freely downloadable shape benchmark especially designed for architectural 3D models. It currently contains 2257 objects from various content providers, including companies specialized on 3D CAD applications. All models are classified according to a scheme developed in close cooperation with architects taking into account their specific requirements regarding design and function. Additionally, we show retrieval results for this benchmark using unsupervised and supervised shape retrieval methods and discuss the specific problems regarding retrieval of architectural 3D models.

The PROBADO project: approach and lessons learned in building a digital library system for heterogeneous non-textual documents

The PROBADO project is a research effort to develop and operate advanced Digital Library support for non-textual documents. The main goal is to contribute to all parts of the Digital Library work flow from content acquisition over indexing to search and presentation. While not limited in terms of supported document types, reference support is developed for classical digital music and 3D architectural models. In this paper, we review the overall goals, approaches taken, and lessons learned so far in a highly integrated effort of university researchers and library experts. We address the problem of technology transfer, aspects of repository compilation, and the problem of inter-domain retrieval. The experiences are relevant for other project efforts in the nontextual Digital Library domain.

Learning the compositional structure of man-made objects for 3D shape retrieval

While approaches based on local features play a more and more important role for 3D shape retrieval, the problems of feature selection and similarity measurement between sets of local features still remain open tasks. Common algorithms usually measure the similarity between two such sets by either establishing feature correspondences or by using Bag-of-Features (BoF) approaches. While establishing correspondences often involves a lot of manually chosen thresholds, BoF approaches can hardly model the spatial structure of the underlying 3D object. In this paper focusing on retrieval of 3D models representing man-made objects, we try to tackle both of these problems. Exploiting the fact that man-made objects usually consist of a small set of certain shape primitives, we propose a feature selection technique that decomposes 3D point clouds into sections that can be represented by a plane, a sphere, a cylinder, a cone, or a torus. We then introduce a probabilistic framework for analyzing and learning the spatial arrangement of the detected shape primitives with respect to training objects belonging to certain categories. The knowledge acquired in this learning process allows for efficient retrieval and classification of new 3D objects. We finally evaluate our algorithm on the recently introduced 3D Architecture Shape Benchmark, which mainly consists of 3D models representing man-made objects.

Demonstration of user interfaces for querying in 3D architectural content in PROBADO3D

The PROBADO project is a research effort to develop Digital Library support for non-textual documents. The main goal is to contribute to all parts of the Digital Library workflow from content acquisition over semi-automatic indexing to search and presentation. PROBADO3D is a part of the PROBADO framework designed to support 3D documents, with a focus on the Architectural domain. This demonstration will present a set of specialized user interfaces that were developed for content-based querying in this document domain.

Enrichment and Preservation of Architectural Knowledge

In the context of the EU FP7 DURAARK project (2013–2016), inter-disciplinary methods, technologies and tools have been researched and developed, that support the Long Term Preservation of semantically enriched digital representations of built structures. The results of the research efforts include approaches of semi-automatically deriving building models from point cloud data sets acquired from laser scans and the integration and overlay of such representations with explicit Building Information Models (BIM). We introduce novel ways for the further semantic enrichment of such hybrid building models with contextual data and vocabularies from external resources using Linked Data (LD) and the recognition relevant features and building components. A special focus of the research reported here lies on strategies and policies for their long term archival, information retrieval based on rich semantic metadata and the use of such archival systems in research and commercial scenarios. We introduce a set of prototypical, open-source tools implementing these features that have been integrated into a modular preservation framework called the “DURAARK Workbench”.

Towards the extraction of hierarchical building descriptions from 3D indoor scans

We present a new method for the hierarchical decomposition of 3D indoor scans and the subsequent generation of an according hierarchical graph-based building descriptor. The hierarchy consists of four basic levels with according entities, building - storey - room - object. All entities are represented as attributed nodes in a graph and are linked to the upper level entity they are located in. Additionally, nodes of the same level are linked depending on their spatial and topological relationship. The hierarchical description enables easy navigation in the formerly unstructured data, measurement takings, as well as carrying out retrieval tasks that incorporate geometric, topological, and also functional building properties describing e.g. the designated use of single rooms according to the objects it contains. In contrast to previous methods which either focus on the segmentation into rooms or on the recognition of indoor objects, our holistic approach incorporates a rather large spectrum of entities on different semantic levels that are inherent to 3D building representations. In our evaluation we show the feasibility of our method for extraction of hierarchical building descriptions for various tasks using synthetic as well as real world data.

Efficient retrieval of 3D building models using embeddings of attributed subgraphs

We present a novel method for retrieval and classification of 3D building models that is tailored to the specific requirements of architects. In contrast to common approaches our algorithm relies on the interior spatial arrangement of rooms instead of exterior geometric shape. We first represent the internal topological building structure by a Room Connectivity Graph (RCG). To enable fast and efficient retrieval and classification with RCGs, we transform the structured graph representation into a vector-based one by introducing a new concept of subgraph embeddings. We provide comprehensive experiments showing that the introduced subgraph embeddings yield superior performance compared to state-of-the-art graph retrieval approaches.