3D Semantic Model Research Articles

Enhancing Heritage fruition through 3D semantic modelling and digital tools: the INCEPTION project

The INCEPTION project, “Inclusive Cultural Heritage in Europe through 3D Semantic Modelling”, started in June 2015 and lasting four years, aims at developing advanced 3D modelling for accessing and understanding European cultural assets. One of the main challenges of the project is to close the gap between effective user experiences of Cultural Heritage via digital tools and representations, and the enrichment of the scientific knowledge. Within this framework, the INCEPTION project goals are consistently aligned while accomplishing the main objectives of accessing, understanding and strengthening European cultural heritage by means of enriched 3D models. At the end of the third year of activity, the project is now facing different challenging actions starting from already developed advancement in 3D data capturing and holistic digital documentation, under interdisciplinary and cross-cutting fields of knowledge. In this direction, the approach and the methodology for semantic organization and data management toward H-BIM modelling will be presented, as well as a preliminary nomenclature for semantic enrichment of heritage 3D models. According to the overall INCEPTION workflow, the H-BIM modelling procedure starts with documenting user needs, including experts and non-experts. The identification of the Cultural Heritage buildings semantic ontology and data structure for information catalogue will allow the integration of semantic attributes with hierarchically and mutually aggregated 3D digital geometric models for management of heritage information.

Structural semantic 3D modeling technology in the integrated design process

The paper presents an approach to computer-aided design based on a bijective correspondence between product structure and the process of developing its digital 3D model. The approach proposed makes it possible to retain the semantic integrity of a design concept and reduce development time.

Semantic as-built 3D modeling of structural elements of buildings based on local concavity and convexity

The aim of this study is to propose a method for generating as-built BIMs from laser-scan data obtained during the construction phase, particularly during ongoing structural works. The proposed method consists of three steps: region-of-interest detection to distinguish the 3D points that are part of the structural elements to be modeled, scene segmentation to partition the 3D points into meaningful parts comprising different types of elements (e.g., floors, columns, walls, girders, beams, and slabs) using local concave and convex properties between structural elements, and volumetric representation. The proposed method was tested in field experiments by acquiring and processing laser-scan data from construction sites. The performance of the proposed method was evaluated by quantitatively measuring how accurately each of the structural elements was recognized as its functional semantics. Overall, 139 elements of the 141 structural elements (99%) in the two construction sites combined were recognized and modeled according to their actual functional semantics. As the experimental results imply, the proposed method can be used for as-built BIMs without any prior information from as-planned models.

INTEGRATED DATA CAPTURING REQUIREMENTS FOR 3D SEMANTIC MODELLING OF CULTURAL HERITAGE: THE INCEPTION PROTOCOL

Abstract. The generation of high quality 3D models can be still very time-consuming and expensive, and the outcome of digital reconstructions is frequently provided in formats that are not interoperable, and therefore cannot be easily accessed. This challenge is even more crucial for complex architectures and large heritage sites, which involve a large amount of data to be acquired, managed and enriched by metadata. In this framework, the ongoing EU funded project INCEPTION – Inclusive Cultural Heritage in Europe through 3D semantic modelling proposes a workflow aimed at the achievements of efficient 3D digitization methods, post-processing tools for an enriched semantic modelling, web-based solutions and applications to ensure a wide access to experts and non-experts. In order to face these challenges and to start solving the issue of the large amount of captured data and time-consuming processes in the production of 3D digital models, an Optimized Data Acquisition Protocol (DAP) has been set up. The purpose is to guide the processes of digitization of cultural heritage, respecting needs, requirements and specificities of cultural assets.

OpenBIM Framework for a Collaborative Historic Preservation System

The authors have developed a novel system framework for a historic preservation system utilising open standards and open source tools. The framework enables the integration of open standard 3D models and GIS in a virtual environment (VE). It also allows the storage and harvesting of data via an open source web-based central repository. The framework is designed for bi-directional interoperability when utilised for monument tracking, preservation, conservation, re-excavation, etc. To date, no such system framework exists for the development and management of historic monuments using open standards e57 for terrestrial laser scanning (TLS) data, the industry foundation classes (IFC) for 3D semantic rich models, and CityGML for integrating IFC and GIS in a VE. The web-based VE is accessed from the BimServer and envisaged for monitoring monument erosion, movement, and damage to monuments by human activity or nature, collaboration between different industry actors to share knowledge, experience, and expertise globally, etc.

Semantic 3D Reconstruction of Furnished Interiors Using Laser Scanning and RFID Technology

AbstractTerrestrial Laser Scanning (TLS) technology is increasingly used for the generation of accurate three-dimensional (3D) models of objects and scenes. But converting the acquired 3D point cloud data into a representative, semantic 3D model of the scene requires advanced processing and skills. This research field is challenging, particularly when considering inhabited, furnished environments that are characterised by clutter and occlusions. This paper presents a TLS data-processing pipeline aimed at producing semantic 3D models of furnished office and home interiors. The structure of rooms (floor, ceiling, and walls with window and door openings) is created using Boundary Representation (B-Rep) models that not only encode the geometry of those elements, but also their connectivity. Windows and doors are recognized and modeled using a novel method based on molding detection. For the furniture, the approach uniquely integrates smart technology [radio frequency identification (RFID)] that is increasingl...

Designing 3D Semantic Model in LOD4 to Simulate Building Utility Network

In recent years, most of the three dimension city models have been introduced as completely graphic or geometric models, and the semantic and topographic aspects of the models have been neglected. However, in addition to graphic aspects, the semantic three dimension models also comprise of ontological structure including thematic classes, attributes and their mutual relations. City GML (Geographical Markup Language) is an international standard to display and exchange threedimension city models. City GML includes five Different Level Of Details (LOD), and in addition to exterior images of the building, it shows the interior space including the rooms, doors, windows, etc. The aim of the present study is to design a semantic model of the buildings in LOD4 with a precision of 20 square centimeters. LOD4 covers the interior space, and gives a comprehensive and accurate view of both interior and exterior spaces of the building. The gas installations of the buildings are also included in the model as a new and practical design to expand the use of City GML. Such models can be used in intelligent management of the buildings, installations management, and also crisis management in interior spaces of the buildings.

Computational geometry in the context of building information modeling

Building energy analysis has gained attention in recent years, as awareness for energy efficiency is rising in order to reduce greenhouse gas emissions. At the same time, the building information modeling paradigm is aiming to develop comprehensive digital representations of building characteristics based on semantic 3D models. Most of the data required for energy performance calculation can be found in such models; however, extracting the relevant data is not a trivial problem. This article presents an algorithm to prepare input data for energy analysis based on building information models. The crucial aspect is geometric simplification according to semantic constraints: the building element geometries are reduced to a set of surfaces representing the thermal shell as well as the internal boundaries. These boundary parts are then associated with material layers and thermally relevant data. The presented approach, previously discussed at the International Academic Conference on Places and Technologies (Ladenhauf et al., 2014), significantly reduces the needed time for energy analysis.

RGB-D Indoor Plane-based 3D-Modeling using Autonomous Robot

Abstract. 3D model of indoor environments provide rich information that can facilitate the disambiguation of different places and increases the familiarization process to any indoor environment for the remote users. In this research work, we describe a system for visual odometry and 3D modeling using information from RGB-D sensor (Camera). The visual odometry method estimates the relative pose of the consecutive RGB-D frames through feature extraction and matching techniques. The pose estimated by visual odometry algorithm is then refined with iterative closest point (ICP) method. The switching technique between ICP and visual odometry in case of no visible features suppresses inconsistency in the final developed map. Finally, we add the loop closure to remove the deviation between first and last frames. In order to have a semantic meaning out of 3D models, the planar patches are segmented from RGB-D point clouds data using region growing technique followed by convex hull method to assign boundaries to the extracted patches. In order to build a final semantic 3D model, the segmented patches are merged using relative pose information obtained from the first step.

Semantic as-built 3d models including shades for the evaluation of solar influence on buildings

This paper presents a working methodology for the automatic generation of as-built Building Information Models (BIMs) including shade surfaces able to be subjected to solar analysis, given the great effect that shading has in the energy consumption of the building. This methodology goes from data acquisition with a laser scanner, to the shape extraction of the building and its surroundings from the acquired data through curvature analysis and Region Growing algorithms, to end with the conversion of the extracted elements, which define both the building and the shading elements, to BIM components. BIM is defined in gbXML language, which was especially designed to be used in energy software, so that the generated BIM with shades can be directly introduced in energy analysis software.

Constraint Solving in History-Independent Semantic 3D Modeling

Constraint Solving in History-Independent Semantic 3D Modeling

Exploiting spatial descriptions in visual scene analysis

The reliable automatic visual recognition of indoor scenes with complex object constellations using only sensor data is a nontrivial problem. In order to improve the construction of an accurate semantic 3D model of an indoor scene, we exploit human-produced verbal descriptions of the relative location of pairs of objects. This requires the ability to deal with different spatial reference frames (RF) that humans use interchangeably. In German, both the intrinsic and relative RF are used frequently, which often leads to ambiguities in referential communication. We assume that there are certain regularities that help in specific contexts. In a first experiment, we investigated how speakers of German describe spatial relationships between different pieces of furniture. This gave us important information about the distribution of the RFs used for furniture-predicate combinations, and by implication also about the preferred spatial predicate. The results of this experiment are compiled into a computational model that extracts partial orderings of spatial arrangements between furniture items from verbal descriptions. In the implemented system, the visual scene is initially scanned by a 3D camera system. From the 3D point cloud, we extract point clusters that suggest the presence of certain furniture objects. We then integrate the partial orderings extracted from the verbal utterances incrementally and cumulatively with the estimated probabilities about the identity and location of objects in the scene, and also estimate the probable orientation of the objects. This allows the system to significantly improve both the accuracy and richness of its visual scene representation.

Benefit of the integration of semantic 3D models in a fire-fighting VR simulator

Virtual reality (VR) simulators have become a great tool for training purposes, especially for risky and uncertain situations such as today’s widely extended driving or flying simulators. One of these cases is the fire-fighting simulators. The usage of a VR simulator to support the training process of fire fighters and managers has two main advantages. On one hand, it supports the simulation of complex scenarios like big cities, where a fire cannot be simulated in the real world; and on the other hand, fire-fighting VR simulators allow trainees to experience situations as close as possible to real fire events, thereby reducing the probability of accidents when they are going through training exercises with real fire. However, the success of the VR simulator depends on how close to reality the simulation process is, one of the most important aspects to ensure how realistic the scenarios shown in the training sessions are. This paper discusses how existing and dynamic 3D geoinformation can be loaded into a fire-fighting VR system and how the preservation of the semantic knowledge can benefit the user experience in the VR simulator. Semantic technologies are intended to help in the selection of information to get a seamless integration between the GIS data, the VR system and the tasks and users involved in the fire-fighting processes. The benefit of semantics is illustrated with some practical cases.

Automatic Construction of 3D Basic-Semantic Models of Inhabited Interiors Using Laser Scanners and RFID Sensors

This paper is focused on the automatic construction of 3D basic-semantic models of inhabited interiors using laser scanners with the help of RFID technologies. This is an innovative approach, in whose field scarce publications exist. The general strategy consists of carrying out a selective and sequential segmentation from the cloud of points by means of different algorithms which depend on the information that the RFID tags provide. The identification of basic elements of the scene, such as walls, floor, ceiling, windows, doors, tables, chairs and cabinets, and the positioning of their corresponding models can then be calculated. The fusion of both technologies thus allows a simplified 3D semantic indoor model to be obtained. This method has been tested in real scenes under difficult clutter and occlusion conditions, and has yielded promising results.

Drawing-Based Procedural Modeling of Chinese Architectures

This paper presents a novel modeling framework to build 3D models of Chinese architectures from elevation drawing. Our algorithm integrates the capability of automatic drawing recognition with powerful procedural modeling to extract production rules from elevation drawing. First, different from the previous symbol-based floor plan recognition, based on the novel concept of repetitive pattern trees, small horizontal repetitive regions of the elevation drawing are clustered in a bottom-up manner to form architectural components with maximum repetition, which collectively serve as building blocks for 3D model generation. Second, to discover the global architectural structure and its components' interdependencies, the components are structured into a shape tree in a top-down subdivision manner and recognized hierarchically at each level of the shape tree based on Markov Random Fields (MRFs). Third, shape grammar rules can be derived to construct 3D semantic model and its possible variations with the help of a 3D component repository. The salient contribution lies in the novel integration of procedural modeling with elevation drawing, with a unique application to Chinese architectures.