3D Indoor Models Research Articles

A voxel-based 3D indoor model to support 3D pedestrian evacuation simulations

Three-dimensional (3D) indoor models enable realistic simulation of pedestrian evacuations in 3D space. However, existing indoor models do not fully identify navigable spaces located above/below physical components (e.g., chairs, stairs, desks), and the connections between these spaces and their navigable surfaces remain underexplored. These limitations hinder the accurate modelling of 3D pedestrian motions during evacuations. To address the shortcomings, we present novel automatic approaches to create a voxel-based 3D indoor model. First, based on different types of 3D motions, indoor spaces are automatically identified as freely navigable spaces, navigable spaces under conditions and non-navigable spaces. Then, navigable surfaces are identified to simulate moving above/below in 3D space. Vertical links are identified to connect the spaces and surfaces at different heights. A 3D pathfinding approach is subsequently developed to validate the model’s effectiveness. Finally, we evaluate the created model’s accuracy, robustness and efficiency across different voxel sizes. The primary findings are: 1) as voxel size increases, accuracy decreases, robustness changes and efficiency improves; 2) with very small voxel sizes, identification becomes challenging, while larger voxel sizes prevent some spaces and vertical links from being identified; and 3) indoor spaces with complex geometries require more time for identification. Our key contributions are: 1) an approach for automatically identifying indoor spaces for 3D motions; 2) two approaches for automatically identifying navigable surfaces and vertical links; and 3) a 3D pathfinding approach considering distinct 3D motion costs. This work advances 3D modelling to facilitate more realistic evacuation simulations.

Voxel Size Performance in 3D Indoor Model Creation for 3D Evacuation Simulations

Abstract. 3D voxel models facilitate simulating 3D pedestrian motions within indoor environments with diverse furniture, equipment, and complex room layouts featuring stairs and ramps. Reasonable voxel sizes enable the effective identifications of indoor spaces, navigable surfaces, and vertical links from voxel models for the simulations. However, the performance of different voxel sizes on these identifications has yet to be thoroughly investigated. The choice of appropriate voxel sizes for various 3D evacuation scenarios has to be clarified to avoid unreliable simulation outcomes. Therefore, this paper aims to assess the performance of different voxel sizes on the identifications of the spaces, surfaces, and links. Firstly, we introduce three key metrics for the assessment: accuracy, robustness, and efficiency. Then, three IFC models are selected based on predefined criteria for the tests. Experimental procedures are elaborated to include data preparation, collection, and analysis. The primary findings of this investigation are: 1) with increasing voxel sizes, accuracy decreases, robustness changes, and efficiency increases; 2) larger indoor spaces exhibit smaller changes in accuracy across different voxel sizes, while smaller indoor spaces show more significant changes in accuracy. 3) when the voxel size is tiny, it becomes challenging to complete identifications. Conversely, with relatively larger voxel sizes, some indoor spaces and vertical links cannot be identified. 4) Indoor spaces with complex and non-orthogonal geometries require more identification time. This work is expected to advance the creation of 3D indoor model to facilitate 3D evacuation simulations.

Parametric indoor 3D reconstruction based on binary image feature point extraction

Abstract In recent years, with the rapid advancement of laser scanning technology, indoor 3D modeling technology has significantly developed, becoming a research hotspot in the field of 3D reconstruction. Addressing the issues of high complexity in decoding point cloud surfaces in traditional indoor reconstruction, this paper proposes an indoor 3D reconstruction method based on binary image feature point extraction. The method utilizes point cloud data for semantic segmentation of the indoor environment to accurately extract the ceiling area, which is then projected onto the XOY plane and converted into a binary image, thereby transforming the 3D problem into a 2D problem. Subsequently, key feature points are identified through feature point detection technology, and systematic model construction is carried out based on the sequence of these feature points and their associated parameters. This method effectively simplifies the complexity of traditional 3D modeling, achieving accurate and parametric indoor 3D reconstruction.

High-resolution topographic surveying and change detection with the iPhone LiDAR.

This paper introduces a comprehensive protocol leveraging open-access techniques to create small- to medium-scale 3D representations of the environment by using iPhone and iPad light detection and ranging (LiDAR). The protocol focuses on two capabilities of the iPhone LiDAR. The first capability is 3D modeling: iPhone LiDAR rapidly generates detailed indoor and outdoor 3D models, providing insights into object size, volume and geometry. The second capability is change detection: the 3D models created by the LiDAR sensor can be used for precise measurement of changes over time. Compared to other 3D topographic surveying methods, this method is rapid, high resolution, low cost and easy to use. The protocol outlines iPhone LiDAR scanning practices, model export and change detection. The expected results after executing the protocol are (i) a detailed 3D model of a small- to medium-sized object or area of interest and (ii) a distance point cloud revealing change between two point clouds of the same object or area between different times. The entire protocol can be conducted within 2 h by anyone with an iPhone with the LiDAR sensor and a computer. This protocol empowers scientists, students and community members conducting research with a cheap, easy-to-use method for addressing a range of questions and challenges, thus benefiting experts and the broader community.

A Review of Datasets for 3D Indoor Modeling

Abstract: Indoor modeling has witnessed a substantial transformation in recent years, driven by advancements in 3D sensing technologies and data processing techniques. Point cloud datasets, representing detailed geometric information of indoor environments, play a pivotal role in enabling various applications such as architectural design, navigation, and augmented reality, and Building Information Modeling (BIM). This paper provides a comprehensive overview and comparision of the stateof-the-art datasets specifically curated for indoor modeling.

A Semantic-Sensitive Approach to Indoor and Outdoor 3D Data Organization

Building model data organization is often programmed to solve a specific problem, resulting in the inability to organize indoor and outdoor 3D scenes in an integrated manner. In this paper, existing building spatial data models are studied, and the characteristics of building information modeling standards (IFC), city geographic modeling language (CityGML), indoor modeling language (IndoorGML), and other models are compared and analyzed. CityGML and IndoorGML models face challenges in satisfying diverse application scenarios and requirements due to limitations in their expression capabilities. It is proposed to combine the semantic information of the model objects to effectively partition and organize the indoor and outdoor spatial 3D model data and to construct the indoor and outdoor data organization mechanism of “chunk-layer-subobject-entrances-area-detail object.” This method is verified by proposing a 3D data organization method for indoor and outdoor space and constructing a 3D visualization system based on it.

Hybrid 3D Reconstruction of Indoor Scenes Integrating Object Recognition

Indoor 3D reconstruction is particularly challenging due to complex scene structures involving object occlusion and overlap. This paper presents a hybrid indoor reconstruction method that segments the room point cloud into internal and external components, and then reconstructs the room shape and the indoor objects in different ways. We segment the room point cloud into internal and external points based on the assumption that the room shapes are composed of some large external planar structures. For the external, we seek for an appropriate combination of intersecting faces to obtain a lightweight polygonal surface model. For the internal, we define a set of features extracted from the internal points and train a classification model based on random forests to recognize and separate indoor objects. Then, the corresponding computer aided design (CAD) models are placed in the target positions of the indoor objects, converting the reconstruction into a model fitting problem. Finally, the indoor objects and room shapes are combined to generate a complete 3D indoor model. The effectiveness of this method is evaluated on point clouds from different indoor scenes with an average fitting error of about 0.11 m, and the performance is validated by extensive comparisons with state-of-the-art methods.

A Motion-Based Conceptual Space Model to Support 3D Evacuation Simulation in Indoor Environments

Three-dimensional (3D) indoor models are a crucial component to simulate pedestrian evacuations realistically in indoor environments. However, existing 3D indoor models cannot fully represent realistic indoor environments to enable the simulation of 3D pedestrian motions in evacuations because spaces above/below some physical components (e.g., desks, chairs) have been largely overlooked. Thus, this paper introduces a conceptual space model to advance a space identification and classification scheme that can fully capture 3D pedestrian motions. This paper first proposes the definition and parameterisation of different 3D pedestrian motions. Then, the definition and specifications of three categories of space components are elaborated on based on the motions. Finally, a voxel-based approach is introduced to identify and classify the space components, which are demonstrated by an illustrative example. This work contributes to advancing 3D indoor modelling to enable a more realistic simulation of 3D pedestrian motions.

3D indoor modeling and game theory based navigation for pre and post COVID-19 situation.

The COVID-19 pandemic has greatly affected human behavior, creating a need for individuals to be more cautious about health and safety protocols. People are becoming more aware of their surroundings and the importance of minimizing the risk of exposure to potential sources of infection. This shift in mindset is particularly important in indoor environments, especially hospitals, where there is a greater risk of virus transmission. The implementation of route planning in these areas, aimed at minimizing interaction and exposure, is crucial for positively influencing individual behavior. Accurate maps of buildings help provide location-based services, prepare for emergencies, and manage infrastructural facilities. There aren't any maps available for most installations, and there are no proven techniques to categorize features within indoor areas to provide location-based services. During a pandemic like COVID-19, the direct connection between the masses is one of the significant preventive steps. Hospitals are the main stakeholders in managing such situations. This study presents a novel method to create an adaptive 3D model of an indoor space to be used for localization and routing purposes. The proposed method infuses LiDAR-based data-driven methodology with a Quantum Geographic Information System (QGIS) model-driven process using game theory. The game theory determines the object localization and optimal path for COVID-19 patients in a real-time scenario using Nash equilibrium. Using the proposed method, comprehensive simulations and model experiments were done using QGIS to identify an optimized route. Dijkstra algorithm is used to determine the path assessment score after obtaining several path plans using dynamic programming. Additionally, Game theory generates path ordering based on the custom scenarios and user preference in the input path. In comparison to other approaches, the suggested way can minimize time and avoid congestion. It is demonstrated that the suggested technique satisfies the actual technical requirements in real-time. As we look forward to the post-COVID era, the tactics and insights gained during the pandemic hold significant value. The techniques used to improve indoor navigation and reduce interpersonal contact within healthcare facilities can be applied to maintain a continued emphasis on safety, hygiene, and effective space management in the long term. The use of three-dimensional (3D) modeling and optimization methodologies in the long-term planning and design of indoor spaces promotes resilience and flexibility, encouraging the adoption of sustainable and safe practices that extend beyond the current pandemic.

Evaluating the Accuracy of UAV and TLS for 3D Indoor Modelling in Large-Scale Building Environments

The accuracy of 3D indoor reconstructed models is critical in various applications such as indoor navigation, virtual reality (VR), and building information modelling (BIM). This research study aims to evaluate the accuracy of Unmanned Aerial Vehicles (UAV) and Terrestrial Laser Scanning (TLS) for 3D indoor modelling in large-scale building environments. To achieve this, several evaluations were made towards the number of point clouds, estimated costs and accuracy of the 3D indoor reconstructed model generated from dense point clouds acquired by Unmanned Aerial Vehicle (UAV) and Terrestrial Laser Scanner (TLS). A small indoor classroom was selected for this study approximately 100m2. In UAV data acquisition, three (3) flight missions were set up at the front, left and right views. Meanwhile, five (5) scanning stations were placed on-site for the TLS method. Due to various different flight mission views in the UAV dataset, the number of point clouds was quite higher compared to the TLS method. However, a better-quality visualization of the TLS model has been obtained as opposed to the UAV 3D model. For the required time to generate a 3D model, it showed that UAV processing time was more consuming lots of time than the TLS method, especially when georeferencing the overlapping photographs. In terms of accuracy, the RMSE value from TLS was better than UAV at 0.003m compared to UAV at 0.021m. Overall, this study provides insights into the accuracy and suitability of UAV and TLS for 3D indoor modelling in large-scale building environments. The results can inform decision-making processes in various industries such as architecture, engineering, and construction, where accurate and reliable 3D models are crucial for design, planning, and management purposes.

Automatic extraction and reconstruction of a 3D wireframe of an indoor scene from semantic point clouds

ABSTRACT Accurate indoor 3D models are essential for building administration and applications in digital city construction and operation. Developing an automatic and accurate method to reconstruct an indoor model with semantics is a challenge in complex indoor environments. Our method focuses on the permanent structure based on a weak Manhattan world assumption, and we propose a pipeline to reconstruct indoor models. First, the proposed method extracts boundary primitives from semantic point clouds, such as floors, walls, ceilings, windows, and doors. The primitives of the building boundary are aligned to generate the boundaries of the indoor scene, which contains the structure of the horizontal plane and height change in the vertical direction. Then, an optimization algorithm is applied to optimize the geometric relationships among all features based on their categories after the classification process. The heights of feature points are captured and optimized according to their neighborhoods. Finally, a 3D wireframe model of the indoor scene is reconstructed based on the 3D feature information. Experiments on three different datasets demonstrate that the proposed method can be used to effectively reconstruct 3D wireframe models of indoor scenes with high accuracy.

Progressive Model-Driven Approach for 3D Modeling of Indoor Spaces

This paper focuses on the 3D modeling of the interior spaces of buildings. Three-dimensional point clouds from laser scanners can be considered the most widely used data for 3D indoor modeling. Therefore, the walls, ceiling and floor are extracted as the main structural fabric and reconstructed. In this paper, a method is presented to tackle the problems related to the data including obstruction, clutter and noise. This method reconstructs indoor space in a model-driven approach using watertight predefined models. Employing the two-step implementation of this process, the algorithm is able to model non-rectangular spaces with an even number of sides. Afterwards, an “improvement” process increases the level of details by modeling the intrusion and protrusion of the model. The 3D model is formed by extrusion from 2D to 3D. The proposed model-driven algorithm is evaluated with four benchmark real-world datasets. The efficacy of the proposed method is proved by the range of [77%, 95%], [85%, 97%] and [1.7 cm, 2.4 cm] values of completeness, correctness and geometric accuracy, respectively.

3D Modelling Building of District Johar Baru Using ArcGIS Pro and CityEngine

Various image-based approaches are currently available for 3 Dimensional (3D) modelling. Among them, two main geomatics techniques are photogrammetry and laser scanning. Owing to the limited availability of photographs to the general populace, the employment of satellite and aerial photogrammetry is rather intricate. In light of Jakarta's absence of a comprehensive three-dimensional portrayal, there is an impetus to generate accurate, precise, and detailed tri-dimensional visualizations of a particular sector within Jakarta, specifically Johar Baru. This endeavor shall be undertaken through a technique termed LoD, with ArcGIS and CityEngine serving as the fundamental instruments for this initiative. To do this, we provided a framework that leverages existing technologies and data sources to reduce the time and work needed for prospective 3D indoor routing applications. Our indoor and building 3D models were created using the CityEngine procedural modelling approach from CAD files and building footprints. This work does not address data delivery or administration; it primarily focuses on 3D visualization. Cadastral, visualization, and non-functional requirements are the three main areas of 3D visualization requirements against which CityEngine was assessed. A case study corresponding to a real issue that has already been recognized in Portugal is used to evaluate the problem. The results are encouraging. The CityEngine won't be the best choice for all users because of the steep learning curve. Using CityEngine, we can show and design a 3D reality in Johar Baru.

MANAGING INDOOR MOVABLE ASSETS IN 3D USING CITYGML FOR SMART CITY APPLICATIONS

Abstract. Smart Cities are the extensive use of the Internet of Things (IoT) and open data to optimise the flow of energy, people, and data, especially in city infrastructure, management, and services. Nowadays, city or building management has become widespread regarding smart city development. Moreover, Asset Management is becoming increasingly important for cities to have the conditions for continuous development, buildings and urban infrastructures that must be planned more efficiently and sustainably. On the other hand, asset management is no longer appropriate since the appearance properties, textures, and materials attached to city models drastically increase the loading time for visualisation and spatial analysis. Besides, different applications or users demand different Level of Details (LODs); hence, a suitable 3D indoor model is needed. However, the available models representing indoor spaces’ geometric and semantic information are rarely described. Thus, to achieve effective asset management, we proposed organising indoor movable assets of indoor building models using CityGML. This paper aims to model the indoor movable assets based on Level of Detail 4 (LOD4) using CityGML concept. There are several limitations of current indoor movable asset management practice, such as the unavailability of asset visualisation, outdated asset information, and a tedious filing system. Therefore, integrating indoor asset management and the CityGML seems to be a better option to manage the 3D asset management information efficiently. This study uses the 3D building model in level 4 (LOD4) to conceptualise and organise the indoor asset information and visualisation. The final output of this study is indoor asset information in LOD 4. Each of the components can be viewed individually together with its information. For future use, the model can be used to serve various smart city applications such as indoor navigation, design, and planning.

Pathfinding method for an indoor drone based on a BIM-semantic model

The application of the drone in construction monitoring, anti-terrorism investigation, and emergency response in large buildings is the development trend of indoor navigation in the future. At present, the indoor three-dimensional (3D) map model based on Building Information Modeling (BIM) is mainly used for indoor pathfinding for pedestrians and ground mobile devices. However, it lacks the spatial description of the vertical dimension, which cannot meet the needs of the navigation application of an indoor drone. Therefore, we propose a pathfinding method for an indoor drone based on a BIM-semantic model. First, the semantic and geometric information in the BIM model is extracted and mapped to voxels to generate an indoor 3D map model called BI3DM. The model subdivides voxel types, which are suitable for subsequent pathfinding for a drone and even pedestrian. Then, a BI3DM-based pathfinding algorithm for an indoor drone is proposed. The algorithm optimizes the path for starting and target locations near the ground, and imports the drone size into the algorithm, which can obtain a more reasonable path. Finally, three types of BIM models are experimented with to verify the effectiveness of the study and its compatibility with pedestrians and ground mobile devices.

Autonomous Exploration of Unknown Indoor Environments for High-Quality Mapping Using Feature-Based RGB-D SLAM

Simultaneous localization and mapping (SLAM) system-based indoor mapping using autonomous mobile robots in unknown environments is crucial for many applications, such as rescue scenarios, utility tunnel monitoring, and indoor 3D modeling. Researchers have proposed various strategies to obtain full coverage while minimizing exploration time; however, mapping quality factors have not been considered. In fact, mapping quality plays a pivotal role in 3D modeling, especially when using low-cost sensors in challenging indoor scenarios. This study proposes a novel exploration algorithm to simultaneously optimize exploration time and mapping quality using a low-cost RGB-D camera. Feature-based RGB-D SLAM is utilized due to its various advantages, such as low computational cost and dense real-time reconstruction ability. Subsequently, our novel exploration strategies consider the mapping quality factors of the RGB-D SLAM system. Exploration time optimization factors are also considered to set a new optimum goal. Furthermore, a Voronoi path planner is adopted for reliable, maximal obstacle clearance and fixed paths. According to the texture level, three exploration strategies are evaluated in three real-world environments. We achieve a significant enhancement in mapping quality and exploration time using our proposed exploration strategies compared to the baseline frontier-based exploration, particularly in a low-texture environment.

IFC2INDOORGML: AN OPEN-SOURCE TOOL FOR GENERATING INDOORGML FROM IFC

Abstract. The interest in 3D indoor models has been continuously growing. Most such models are made available as point clouds or BIM (e.g., IFC), the former being generally provided as unstructured information while the latter comes highly structured and rich in semantic information. IFC models are consequently more suitable for direct use, but they can be very complex and contain too many details, which often raises privacy concerns. IndoorGML is one of the standards for describing 3D indoor space with the purpose of supporting Location Based Services (LBS). It relies on solid scientific concepts and offers a high flexibility with extension mechanisms. It provides a geometric, topological, and semantic description of the indoor which facilitates specifically applications like indoor navigation or facility management. Additionally, it can represent complex indoor environments without compromising privacy, thanks to its high level of abstraction. However, despite its solid conceptual basis, IndoorGML is suffering from a lack of practical tools and remains hard to produce, making it largely unavailable. In this project, we developed an open-source tool named ifc2indoorgml allowing to automatically generate IndoorGML models from IFC data. We discuss the workflow and the different development approaches. By making such tool available to the wider public, we expect more 3D IndoorGML models to be created and made freely available for research and development within the spatial community and beyond.

GeoRec: Geometry-enhanced semantic 3D reconstruction of RGB-D indoor scenes

Semantic indoor 3D modeling with multi-task deep neural networks is an efficient and low-cost way for reconstructing an indoor scene with geometrically complete room structure and semantic 3D individuals. Challenged by the complexity and clutter of indoor scenarios, the semantic reconstruction quality of current methods is still limited by the insufficient exploration and learning of 3D geometry information. To this end, this paper proposes an end-to-end multi-task neural network for geometry-enhanced semantic 3D reconstruction of RGB-D indoor scenes (termed as GeoRec). In the proposed GeoRec, we build a geometry extractor that can effectively learn geometry-enhanced feature representation from depth data, to improve the estimation accuracy of layout, camera pose and 3D object bounding boxes. We also introduce a novel object mesh generator that strengthens the reconstruction robustness of GeoRec to indoor occlusion with geometry-enhanced implicit shape embedding. With the parsed scene semantics and geometries, the proposed GeoRec reconstructs an indoor scene by placing reconstructed object mesh models with 3D object detection results in the estimated layout cuboid. Extensive experiments conducted on two benchmark datasets show that the proposed GeoRec yields outstanding performance with 5.19×10-3 mean chamfer distance error for object reconstruction on the challenging Pix3D dataset, 70.45% mAP for 3D object detection and 77.1% 3D mIoU for layout estimation on the commonly-used SUN RGB-D dataset. Especially, the mesh reconstruction sub-network of GeoRec trained on Pix3D can be directly transferred to SUN RGB-D without any fine-tuning, manifesting a high generalization ability.

Volumetric Representation and Sphere Packing of Indoor Space for Three-Dimensional Room Segmentation

Room segmentation is a basic task for the semantic enrichment of point clouds. Recent studies have mainly projected single-floor point clouds to binary images to realize two-dimensional room segmentation. However, these methods have difficulty solving semantic segmentation problems in complex 3D indoor environments, including cross-floor spaces and rooms inside rooms; this is the bottleneck of indoor 3D modeling for non-Manhattan worlds. To make full use of the abundant geometric and spatial structure information in 3D space, a novel 3D room segmentation method that realizes room segmentation directly in 3D space is proposed in this study. The method utilizes volumetric representation based on a VDB data structure and packs an indoor space with a set of compact spheres to form rooms as separated connected components. Experimental results on different types of indoor point cloud datasets demonstrate the efficiency of the proposed method.

MHIBS-Net: Multiscale hierarchical network for indoor building structure point clouds semantic segmentation

Efficient and accurate segmentation of indoor building structure (IBS) from complex indoor scenes is the primary task of indoor 3D modeling. At present, the research on semantic segmentation of indoor scenes mainly focuses on multiple types of objects, while there are few studies on semantic segmentation of IBS. In this paper, we introduce an efficient and lightweight multiscale hierarchical network MHIBS-Net for semantic segmentation of IBS point clouds. On the one hand, normal vector information (NVI) is added and a relative position encoding (RPE) unit is designed in feature encoding (FE) to better capture the local geometric structure information of the IBS point clouds. On the other hand, the traditional feature decoding (FD) process is improved by a newly designed multilevel hierarchy FD method. In this way, the MHIBS-Net can capture the local and global features of IBS point clouds comprehensively so as to realize automatic and efficient semantic segmentation of IBS point clouds. Finally, several groups of experiments are designed to compare and analyze the proposed MHIBS-Net with some classical semantic segmentation networks, and the Stanford large-scale 3D indoor spaces (S3DIS) dataset is used for experimental verification. Experimental results show that MHIBS-Net can achieve high-precision and high-efficiency automatic semantic segmentation of IBS point clouds in complex indoor environments.