Modeling Standards and File Formats for Indoor Mapping
Jorge Chen and Keith C. Clarke
Department of Geography, University of California, Santa Barbara, 1721 Ellison Hall, Santa Barbara, CA 93106, U.S.A.
Keywords: Indoor Cartography, Indoor Mapping and Modeling, Building Information Modeling, Geographic
Information Systems, Computer-Aided Design, Architecture.
Abstract: Indoor maps provide abstractions of the physical spaces where we spend most of our lives. General purpose
indoor maps have historically taken the form of two-dimensional floor plans, commonly found in public
venues such as shopping malls and cruise ships. Until recently, innovation and development of indoor maps
have remained confined to urban planning and the building industry. Recent interest in indoor mapping for
other applications has extended indoor mapping to 3D and to other domains, with a growing emphasis on
commerce and general wayfinding. This paper reviews prevailing modeling standards and file formats
relevant to the modeling and visualization of indoor spaces with the goal of assisting researchers and
developers with finding appropriate formats for indoor modeling and visualization.
1 INTRODUCTION
The absence of positioning and reality capture
systems for indoor spaces has historically limited
indoor mapping to the two domains of urban
planning and the building construction industry.
However, recent advancements in indoor positioning
systems and reality capture technologies and a
maturing market for outdoor mapping and
navigation aids have catalyzed the mapping of
indoor features (Zlatanova et al., 2013). Sample
applications include the eeGeo 3D indoor mapping
platform, MediNav by Connexient for hospital
wayfinding, and Indoor Atlas and Google Maps for
commerce and general wayfinding. Growing interest
in gamification for virtual training also points to
other areas of future growth for research and
development in indoor maps (Muller et al., 2004;
Popescu and Mudure, 2008; Spicer et al., 2016).
We use the term indoor mapping for both the
geometric and semantic abstractions of physical
indoor spaces, defined as inhabitable spaces with
overhead obstructions, e.g., building interiors, ship
interiors, and caves. Reality capture technologies
produce meshes or point clouds—valuable data for
maps but not necessarily maps themselves. A map is
a model that further abstracts the raw reality-capture
data into geometric shapes (i.e., symbols) with both
topology and semantic meaning.
This review examines the three big players in
indoor mapping (CityGML, IFC, and IndoorGML)
that provide data models, semantic frameworks, and
file formats for working with buildings and indoor
data. We also examine other file formats from
computer graphics that have potential relevance to
indoor mapping.
2 CRITERIA FOR REVIEW
We used four criteria for our review: geometric
abstraction, semantic support, spatial referencing
method, and level-of-detail (LOD) support.
Geometric abstraction is the type of geometry used
to represent building features, while semantic
support describes the ability of a model or format to
support higher-level abstractions of the entities
being modeled (Hammer & McLeod, 1981; Mayer
et al., 2013; Stephens, 2008). Spatial referencing is
the method used for describing model measurements
and LOD is a standard or file format’s way of
managing geometric complexity.
Most modeling standards and file formats follow
a handful of approaches for representing geometry
(Mayer et al., 2013; Molenaar, 1989;). 2D
representation uses points, lines, and areas to
symbolize entities. 2.5D representation uses height
or depth values added to each 2D horizontal
coordinate pair to produce a surface. 2.5D surfaces
with equally-spaced points are called grids or
268
Chen, J. and Clarke, K.
Modeling Standards and File Formats for Indoor Mapping.
DOI: 10.5220/0006364202680275
In Proceedings of the 3rd International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2017), pages 268-275
ISBN: 978-989-758-252-3
Copyright © 2017 by SCITEPRESS – Science and Technology Publications, Lda. All rights reserved

rasters, while surfaces with irregularly spaced and
connected points are called triangulated irregular
networks (TIN).
Mayer et al. (2013) listed six different 3D
geometric representations. Voxels are extensions of
rasters in 3D. In cell decomposition, geometric
objects are formed by a Boolean union of geometric
primitives. For constructive solid geometry (CSG)
objects are formed by merging or subtracting
primitive objects using Boolean operations.
Boundary representation (BRep) uses objects
defined by connected faces, edges, and vertices at
the boundary of the solid object. In a parametric
description, objects are defined by numerical
parameters in the form of constraints and
dimensions. Lastly, a tessellation is an extension of
TINs in 3D.
Semantics help associate geometries with entities
in the physical world, such as building parts or
furniture (Billen et al., 2014). Under the semantic
object modeling framework, semantic objects
represent real entities using levels of attributes:
simple, group, and object (Stephens, 2008).
Spatial referencing uses descriptors to define the
locations of objects in physical space. The
International Standards Organization (ISO) and
Open Geospatial Consortium (OGC) identify two
approaches to spatial referencing: by identifier and
by coordinates (ISO, 2002, 2003; OGC, 2010).
Referencing by coordinates uses measurements to
define location, while referencing by identifier uses
geocodes as surrogates for coordinates. Street
addresses, building numbers, and room numbers are
examples of identifiers.
Referencing by coordinates requires both a
measurement system and a measurement starting
point (Bernhardsen, 2002; ISO, 2002; OGC, 2010;
Van Sickle, 2010). A coordinate system defines
properties of the measurement system while a datum
defines the measurement starting point. The
combination of a coordinate system and a datum
forms a unique coordinate reference system (CRS),
which often assumes the classification of the datum.
Earth-based datums can be engineering, geodetic, or
vertical. An engineering datum has local geographic
scope and uses local physical markers as the starting
point. A geodetic datum uses Earth itself as the
starting point, usually Earth’s center or the surface
of an ellipsoidal. A vertical datum uses a gravity-
based geoid to approximate mean sea level for
vertical measurements. Engineering CRSs are nearly
always planar and use linear units of measurement,
which does not pose a problem for most buildings
and indoor spaces. In contrast, geodetic CRSs can
use the angular units of an ellipsoid or the linear
units of a planar projection. Various CRS
transformations exist to convert one CRS to another.
Level of detail (LOD) describes the different
ways a geometric object can be represented at
different levels of generalization (Pham et al., 2015;
Tolmer et al., 2013). The computer graphics
community uses varying polygon counts to describe
LOD, while architects use the term level of
development—LODt here for clarity—to show
representations of architectural objects at various
stages of design and construction. In cartography
and GIS, LOD can imply both differences in
geometric abstractions and in object semantics.
3 MODELING STANDARDS
IFC (Industry Foundation Classes), CityGML (City
Geography Markup Language), and IndoorGML
(Indoor Geography Markup Language) are three
major international standards with relevant guidance
for the 3D modeling of indoor spaces. Each standard
meets the unique needs of a specific application—
IFC for building construction and management,
CityGML for urban modeling, and IndoorGML for
indoor navigation—but they can interoperate. While
the theory behind building modeling has existed
since the 1960s and parallels GIS development,
standardization of building models and modeling
processes is recent—IFC in 2000, CityGML in 2008,
and IndoorGML in 2014.
IFC represents both a standard data exchange
schema for building information modeling (BIM)
data and a file format used by the architecture,
engineering, and construction (AEC) and facilities
management (FM) industries. BIM software such as
Autodesk Revit and Graphisoft ARCHICAD use
their own proprietary formats based on IFC but can
export to IFC format. IFC formats are the ISO STEP
file structure (.ifc), a text file formatted using
extensible markup language, XML (.ifcXML), or a
compressed file (.ifcZIP) containing a .ifc
or .ifcXML file.
As a modeling standard, IFC serves as the
centerpiece of three standards that make up
openBIM, the de facto international BIM
framework; the other two are the buildingSmart Data
Dictionary (bSDD) and information delivery
manuals (IDMs).
IFC mainly supports CSG and sweep geometries,
but it can also support BRep and tessellation. While
IFC supports parametric models, incomplete
implementation by BIM software may make it
Modeling Standards and File Formats for Indoor Mapping
269

necessary to convert proprietary parametric BIM
models into CSG, sweep, or BRep for IFC (Ji et al.,
2011). IFC provides strong support for semantics by
offering true object-based modeling with class
definitions and inheritance. The bSDD serves as the
multi-lingual data dictionary to provide ontologies
and attributes of IFC objects, such as doors, walls,
and structural elements (Petrie, 2016). For spatial
referencing, IFC uses an engineering CRS with an
option to place a building site on a geodetic CRS
using geographic coordinates, an elevation, and,
optionally, the direction of true north using OGC
CRS guidelines (buildingSMART, 2013).
IFC uses the American Institute of Architects
(AIA) version of level of detail called level of
development (LODt) (Reinhardt & Bedrick, 2016).
LODt is the refinement level of building elements at
various stages of design and construction. IFC uses
the five LODts in AIA protocol G202-2013 (LODt
100, 200, 300, 400, and 500) and a sixth for
openBIM (LODt 350), with 100 as the least refined
and 500 as actual construction. Refinement level
may not correspond to geometric detail; detailed
models of building elements can be used in LODt
200 even if they are only placeholders.
CityGML serves as the official standard for
urban modeling within the Open Geospatial
Consortium (OGC) framework, which addresses the
modeling of all things geospatial. Specifically, it
exists as a specialized application schema of the
base-level OGC Geography Markup Language
(GML). Similar to IFC, CityGML represents both a
modeling standard and a file format. As a file
format, CityGML uses an XML schema as defined
in the standard. As a standard, CityGML provides
the foundational framework for urban modeling
upon which more specialized applications—called
application domain extensions or ADEs—can be
built. Examples of CityGML ADEs include
applications for noise, hydrology, solar irradiance,
energy, utilities, and other areas, as well as
integration with IFC. CityGML exclusively uses
BRep geometries, provides the framework for rich
object-based semantics (especially with the use of
ADEs), can accommodate a wide variety of
engineering and geodetic CRSs, and has five levels
of detail (Buyukaslih et al. 2013; Kolbe et al., 2005;
OGC, 2012). For an engineering CRS, the CityGML
2.0 specifications recommend identifying an anchor
point based on a geodetic CRS (OGC, 2012).
LODs for CityGML closely follow the
cartographic concept of LOD, in which an object’s
geometry and semantics (features) are kept separate.
CityGML has five LODs from LOD0 (most
generalized) to LOD4 (most detailed). Since features
and geometry are separated, a single feature can
have different shapes at different LODs and even
have different appearances based on theme. LOD4
has particular relevance to indoor cartography since
it provides the only LOD that reveals indoor
features; building interiors remain void and empty
from LOD0 to LOD3.
IndoorGML was developed solely to support
navigation within indoor environments (Kim, Yoo,
& Li, 2014; Nagel et al., 2010). As a topological
modeling standard, IndoorGML places a strong
emphasis on semantic objects and their topological
relationships and it uses non-overlapping cells in
lieu of geometric representations to model indoor
spaces. These object-based cells need neither
dimension nor spatial location, although the standard
supports adding either or both as an option. As such,
IndoorGML supports both engineering and geodetic
CRSs. Cells can have topological relationships and
can relate to geometry by linking to CityGML or
IFC. The standard does not require geometry-based
LODs, but uses multi-layering to represent different
uses of the same space. One layer may represent
pedestrian travel while another may represent
wireless internet coverage.
Shapefiles require files stored in the same
directory: feature geometries or shapes (.shp), shape
index (.shx), and attribute tables (.dbf). They can
include optional files, such as the projection file
(.prj) for storing for CRS. Shapefiles represent real-
world entities using simple 2D and 3D vector
geometries based on points, lines, areas, and
tessellation (LOC, 2011a). Shapefile features have
no topological capabilities, have basic semantic
capabilities through the use of attributes, can use
engineering or geodetic CRSs through the .prj file,
but have no built-in capabilities for defining LOD.
For 3D representation, shapefiles use multipatch
geometry, a type of BRep (Esri, 2008).
Geodatabases represent real world entities using
semantic objects called features, which allow for
object-based development using feature classes. In
terms of geometry, the geodatabase format uses the
same vector format as shapefiles—i.e., points, lines,
areas, and 3D multipatches—with additional support
for rasters in 2D (LOC, 2011b). The format
currently does not support voxels (Shephard, 2015).
For spatial referencing, the geodatabase supports
both engineering and local CRSs as well as
referencing by identifiers. Geodatabases support
different user-defined LODs due to the separation of
features and geometry. There are three
implementations of the geodatabase—personal, file,
GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management
270

and enterprise—with increasing features and
capabilities (Esri, 2016). Personal geodatabases use
Microsoft Access data files to store data, file
geodatabases use a directory-base file structure, and
enterprise geodatabases use servers.
The open source Facilities Information Spatial
Data Model (FISDM) emerged in 2014 from the
Building Interior Space Data Model (BISDM)
started by Esri and other organizations in 2007 (Rich
& Smith, 2014). FISDM is not a standard but a best
practices framework for CAD-GIS integration built
on the geodatabase format for fusing GIS data with
CityGML, IFC, and other formats. While BISDM
focused exclusively on interior spaces, FISDM
extends those capabilities to include exterior spaces
and the outside built environment. FISDM has the
same 2D and 3D capabilities of its underlying
geodatabase file format.
4 OTHER GRAPHIC FORMATS
Many graphic formats exist for the production,
transmission, or presentation of 2D and 3D building
geometry. McHenry and Bajcsy (2008) identified
140 formats for 3D graphics alone. Data conversion
software Safe FME lists over 250 formats: 197 2D
vector-based formats, some of which also support
3D, with 39 for 3D graphics; 68 2D raster formats;
and 13 point cloud or voxel formats (Safe Software,
2017). Software that can handle IFC, CityGML, or
GIS data can also export graphics into one or more
these formats; conversion software such as Safe
FME can also perform conversions. Here we
examine some key formats relevant to building
modeling and indoor modeling, with an emphasis on
data sources, data exchange, and visualization.
Drawings produced using CAD software
document the design, construction, and modification
of buildings and are excellent sources of data for
indoor mapping. Although BIM has significant
advantages over CAD, industry-wide adoption of
BIM only began less than a decade ago and is not
yet universal (McGraw-Hill Construction, 2012). 2D
and 3D CAD remain widely used in AEC and most
legacy drawings prior to the advent of BIM only
exist in CAD format or hard copy (Business
Advantage, 2016; Coumans, 2017) Prevailing CAD
formats include DXF, DWG, and DGN.
Despite having been around since 1982, the
Autodesk DXF exchange format remains popular
and widely used due to its semi-public nature,
simple structure, and high level of software
compatibility (LOC, 2016b). The open portion of
DXF supports points, lines (including curves), and
areas in 2D/3D, as well as tessellation in 3D; but the
“3D Solid” capability is proprietary to Autodesk and
cannot be edited. DXF uses an engineering CRS and
has support for neither semantics nor LOD, although
Zlatanova et al. (2012) noted that layers can simulate
semantic objects. Note that the terms objects and
entities in the DXF specifications refer to graphical
features (e.g., points and lines) rather than physical
entities (e.g., walls and doors) (Autodesk, 2011).
DWG is Autodesk’s native CAD format for its
AutoCAD software and shares the same properties
as DXF with explicit support for tessellation
(meshes) and BRep with sweeping and extrusion
capabilities (LOC, 2016a).
The DGN format is Bentley’s native CAD
format for its Microstation software and supports
both engineering and geodetic CRSs, with no LOD
capabilities and no semantics, but supports points,
lines (including curves), and areas in 2D/3D and 3D
solids through extrusion in BRep (Bentley Systems,
2016). Unlike DWG/DXF, DGN does not support
tessellation/TINs.
COLLADA is an XML-based ISO-adopted
standard for the exchange of 3D digital assets among
different interactive 3D software applications
(Barnes & Finch, 2008; Khronos Group, 2017). It
uses an object-based approach, has BRep geometry
with limited support for tessellation, supports
custom object-based semantics insofar as they relate
to geometry, and supports multiple LODs.
COLLADA supports use of a local engineering CRS
and one geodetic plus vertical CRS (WGS84 and
WGS84-EGM96).
Web browsers provide universally accessible
platforms for viewing and interacting with building
models and indoor maps. X3D and SVG represent
open XML-based standards for the delivery of 3D
and 2D vector content through HTML5 compliant
web browsers. X3D is a family of open ISO
standards for representing and exchanging 3D
scenes and objects managed by the Web3D
Consortium (Web3D Consortium, 2017). X3D
emerged from an earlier standard called the Virtual
Reality Modeling Language (VRML), which had a
geographic version (GeoVRML). While X3D has no
native capability for custom attributes, it does allow
the linking of attributes in other XML files for that
capability (Geroimenko & Geroimenko, 2006). A
quasi-object-based semantic capability called
grouping allows for the definition and re-use of
constructed geometries, called groups. Other
properties of X3D include multiple LODs, as well as
support for both local engineering and a few
Modeling Standards and File Formats for Indoor Mapping
271

geodetic CRSs (Web3D Consortium, 2008). X3D
data can be presented in XML, Javascript, and Java.
X3D has many other graphical capabilities. Special
capabilities include the ability to handle voxel data
and the ability to support CAD models, including the
use of BRep geometries (Brutzman, 2012).
SVG provides XML-based 2D graphics that are
viewable with web browsers. The latest version, 1.1,
includes support for lines and areas (i.e., rectangle,
circle, ellipse, and polygon) as well as rasters;
however, SVG currently does not support point
geometry. A draft recommendation for version 2
adds a mesh geometry that allows for shape
distortions, but it still omits point geometry. As with
X3D, SVG allows the definition of re-usable groups
and allows linking to other XML files for custom
attributes, although not a native capability (Adams,
2005; Geroimenko & Geroimenko, 2006). It natively
supports engineering CRSs and has a flexible system
for specifying geodetic CRSs using one of three
methods: a web-based uniform resource identifier
(URI), a well-known CRS identifier, or directly
defining the CRS within the xml document. As with
X3D, SVG also supports multiple-representation
LODs (Chang et al., 2004).
KML is an XML-based OGC international
standard for providing 2D and 3D geographic
visualizations in online mapping and virtual globe
browsers. KML supports point, line, and polygon
geometry and 3D geometry using BRep (Isikdag &
Zlatanova, 2010). The KML standard provides a
limited ability to add user-defined attributes to
features (Google, 2013). KML exclusively uses the
World Geodetic System of 1984 (WGS84) for
horizontal coordinates and Earth Gravitational
Model 1996 based on the WGS84 ellipsoid for
elevations (WGS84 EGM96 Geoid). It also supports
user-defined levels of detail through the use of
regions (Burggraf, 2015).
PDF with 3D content provides a convenient
format for delivering 3D models using widely
available PDF readers. 3D models reside in external
files (in one of only two formats, U3D or PRC)
embedded into the PDF document. U3D supports the
visualization of geometric models with attached
attributes, but lacks object-based capabilities such as
inheritance. U3D only supports the use of
tessellation for 3D geometry, uses a local
engineering CRS, and its implementation of LOD
uses reduced polygon counts instead of multiple
representations (Klawonn, 2012). The ISO-adopted
PRC format supports geometry with attached
attributes but falls short of being truly object-based.
In addition to tessellation, PRC also supports BRep
and other geometric representations (PDF3D, 2015).
PRC natively uses a local engineering CRS but some
PDF implementations can provide software-based
transformations to a geodetic CRS.
Table 1 provides a summary of the modeling
standards and file formats covered in this review.
SVG is the only item on this list that cannot support
3D geometry. Items that have points/lines/areas in
addition to 3D geometry denote a format that can
support 2D as well as 3D geometry.
5 DISCUSSION
We divide our discussion into three areas: data
sourcing, software accessibility, and modeling
capabilities. Data sourcing examines data import,
software compatibility examines practical issues of
working with and sharing indoor maps, and
modeling capabilities assesses capabilities and
potential limitations.
Existing building designs serve as excellent
sources of data for creating indoor maps. Among the
reviewed models and formats, BIM models based on
IFC have the greatest potential use for 2D and 3D
indoor mapping due to their very high level of detail,
native 3D geometry, and rich semantics. However,
BIM models only exist for newer structures natively
designed in BIM or for the few older structures
documented in BIM after the fact. For older
structures, CAD drawings provide the next best data
source. If CAD drawings do not exist, then available
hard copy drawings can be used, either through a
scan-to-CAD or scan-to-BIM process or through
manual transcription. Most CAD and hard copy
drawings only exist as 2D line drawings, which
require manual interpretation to convert to 3D.
While CityGML has LOD4 for indoor modeling,
few CityGML models exist at that level of detail. Of
the 15 public urban models listed on the official
CityGML website, none uses LOD4 and only one
uses LOD3. This may be due to a current lack of use
cases for indoor maps—the subject of our research.
Meanwhile, BIM will continue to serve as the most
reliable source of detailed indoor data due to a
commercial need in the AEC/FM industry.
Reality capture takes 3D measurements of indoor
spaces using laser scanning and photogrammetry
and delivers point clouds—massive collections of
points with optional attributes such as color or
intensity. Most BIM and CAD software support
importing point clouds for transforming into their
respective formats.
GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management
272

Table 1: Summary of standards and file formats.
Supporte
d
Geometr
y
Semantic Support Spatial Referencing Level of Detail … Indoor / Total
IFC
6/6
CityGML
1/5
IndoorGML
N/A None N/A
Shapefile
None N/A
Geodatabase
User
FISDM
See geodatabase
DXF/DWG
None N/A
DGN
None N/A
COLLADA
User
X3D
User
SVG
User
KML
User
PDF (U3D)
User
PDF (PRC)
User
All file formats covered in this review, except DWG
and DGN, are open specification formats and have
high levels of compatibility and accessibility using
various software. An open specification format
allows software makers to incorporate it into their
products for cross-platform exchanges. Closed
formats such as DWG, DGN, and RVT only work
with proprietary programs—in this case, AutoCAD,
Microstation, and Revit.
IFC, DXF, and COLLADA are open exchange
formats designed to maximize compatibility and
minimize conflicts when transferring between
different software programs. While software makers
often accommodate these exchange formats, data
loss can still occur. For example, incomplete
integration of parametric geometries between Revit
and IFC can sometimes cause Revit to convert
parametric geometries into BRep or tesselation,
resulting in a loss of geometric parameters.
Of the remaining formats, CityGML, X3D, and
SVG hold significant value for indoor mapping.
CityGML provides the greatest level of support for
the diverse requirements of indoor mapping. It
supports rich semantics, can accommodate both
engineering and geodetic CRSs, and supports the
conventional concept of LOD as used in the
mapping sciences. Additionally, application domain
extensions allow CityGML to be extended for
indoor mapping. While IFC provides greater
flexibility with geometry and offers even richer
semantics support, its lack of robust CRS support
and use of LODt make it too restrictive for indoor
mapping.
A shortcoming of CityGML is its lack of
multiple LODs for indoor spaces. Use of the single
LOD4 provides an all-or-nothing approach in which
LOD4 shows everything indoors while LOD3 shows
the hollow shell of a building. Providing multiple
LODs for indoor spaces can allow various levels of
generalization to occur, e.g., generalizations of
rooms or entire floors.
While CityGML provides a viable framework for
generating and storing indoor maps, it is not
optimized for visualization. For that capability,
CityGML can export to X3D and SVG formats,
which work with most modern web browsers, or to
prevailing 2D GIS web mapping formats. X3D holds
significant promise for the display of 3D indoor
maps. X3D supports rich semantics, engineering and
geodetic CRSs (though limited), and cartographic
LODs. It can also display voxels, which may
provide an alternative form of visualization, as
popularized by the video game Minecraft. While
SVG holds some promise for 2D visualization,
mature GIS products that use shapefiles or
geodatabases may provide a more practical solution.
Using a GIS system for 2D indoor maps can allow
the indoor maps to seamlessly integrate with outdoor
maps, benefit from GIS functionality, and take
advantage of the existing infrastructure already in
Modeling Standards and File Formats for Indoor Mapping
273

place for web-based mapping. FISDM was
developed for this purpose.
6 CONCLUSIONS
Indoor mapping can provide significant economic
value due to the amount of time people spend
indoors. While a multitude of formats exist for
modeling and visualizing indoor spaces, certain
formats can make indoor map development and
presentation more effective based on the specific
needs of a process. Potential areas of continued
development to support indoor mapping include
better integration of indoor and outdoor CRSs, a
more refined concept of LOD for indoors, and the
use of voxels as alternative representations for
viewing by the general public.
ACKNOWLEDGEMENTS
Support for this research has been provided by the
U.S. National Geospatial-Intelligence Agency's
NURI Program.
REFERENCES
Adams, T. (2005). Using SVG and XSLT to Display
Visually Geo-referenced XML. In V. G. Ds. MSc &
C. C. BSc MSc (Eds.), Visualizing Information Using
SVG and X3D (pp. 256–265). Springer London.
Autodesk. (2011, February). AutoCAD 2012 DXF
Reference. Autodesk.
Barnes, M., & Finch, E. L. (2008). COLLADA – Digital
Asset Schema Release 1.5.0 Specification. The
Khronos Group Inc.
Bentley Systems. (2016). MicroStation Wiki. Retrieved 3-
10-2017 from communities.bentley. com/products/
microstation/w/microstation__wiki.
Bernhardsen, T. (2002). Geographic Information Systems:
An Introduction, 3rd Edition. John Wiley & Sons.
Billen, R., Cutting-Decelle, A.-F., Marina, O., de Almeida,
J.-P., M., C., Falquet, G., … Zlatanova, S. (2014). 3D
City Models and urban information: Current issues
and perspectives: European COST Action TU0801. In
R. Billen, A.-F. Cutting-Decelle, O. Marina, J.-P. de
Almeida, C. M., G. Falquet, … S. Zlatanova (Eds.) (p.
I-118). EDP Sciences.
Brutzman, D. (2012, November). X3D Graphics Support
for Computer Aided Design (CAD).
buildingSMART. (2013). 8.18.3.2 IfcCoordinateReference
System. Retrieved March 9, 2017, from
http://www.buildingsmart-tech.org/ifc/IFC4/final/
html/link/ifccoordinatereferencesystem.htm.
Burggraf, D. (2015, August 4). OGC KML 2.3
[Implementation Standard]. Retrieved January 28,
2017, from http://docs.opengeospatial.org/is/12-
007r2/12-007r2.html.
Business Advantage. (2016). 3D Modelling, 2D Drafting
and Simulation CAE (CAD Trends 2016).
www.business-advantage.com/blog/cad-trends-2016-
3d-modelling-2d-drafting-and-simulation-cae/
Buyukaslih, I., Isikdag, U., & Zlatanova, S. (2013).
Exploring the processes of generating LOD (0-2)
CityGML models in greater municipality of Istanbul.
8th 3DGeoInfo Conference & WG II/2 Workshop,
Istanbul, Turkey, 27–29 November 2013, ISPRS
Archives Volume II-2/W1.
Chang, Y.-H., Chuang, T.-R., & Wang, H.-C. (2004).
Adaptive Level-of-Detail in SVG. In 3rd Annual
Conference on Scalable Vector Graphics. Tokyo.
Coumans, F. (2017, February 22). Data Sharing is Smart:
Geo-related Trends in Engineering & Construction.
GIM Magazine.
Esri. (2008, December). The Multipatch Geometry Type.
Esri. Retrieved from https://www.esri.com/library/
whitepapers/pdfs/multipatch-geometry-type.pdf.
Esri. (2016). Types of geodatabases. Retrieved 1-28-2017,
from desktop.arcgis.com/en/arcmap/10.3/manage-
data/geodatabases/types-of-geodatabases.htm.
Geroimenko, V., & Geroimenko, L. (2006). SVG and
X3D: New XML Technologies for 2D and 3D
Visualization. In V. Geroimenko & C. Chen (Eds.),
Visualizing the Semantic Web 124–133. Springer
London.
Google. (2013, November 14). Adding Custom Data.
Retrieved 3-13-2017, from developers.
google.com/kml/documentation/extendeddata.
Hammer, M., & McLeod, D. (1981). Database Description
with SDM: A Semantic Database Model. ACM
Transactions on Database Systems, 6(3), 351–386.
Isikdag, U., & Zlatanova, S. (2010). Interactive modelling
of buildings in Google Earth: A 3D tool for Urban
Planning. In T. Neutens & P. Maeyer (Eds.),
Developments in 3D Geo-Information Sciences (pp.
52–70). Springer Berlin Heidelberg.
ISO. (2002). Geographic information — Spatial
referencing by coordinates (ISO 19111) (No.
ISO/FDIS 19111:2002(E)).
ISO. (2003). Geographic information — Spatial
referencing by geographic identifiers (ISO 19112)
(No. ISO/FDIS 19112:2003(E)).
Ji, Y., Beetz, J., Bonsma, P., Bisbet, N., Katz, C., &
Borrmann, A. (2011). Integration of parametric
geometry into IFC-Bridge. Presented at the Proc. of
the 23th Forum Bauinformatik, Cork, Ireland.
Khronos Group. (2017). COLLADA - 3D Asset Exchange
Schema. Retrieved 1-28-2017, from
www.khronos.org/api/collada/
Kim, J.-S., Yoo, S.-J., & Li, K.-J. (2014). Integrating
IndoorGML and CityGML for Indoor Space. In D.
Pfoser & K.-J. Li (Eds.), Web and Wireless
Geographical Information Systems (pp. 184–196).
Springer Berlin Heidelberg.
GISTAM 2017 - 3rd International Conference on Geographical Information Systems Theory, Applications and Management
274

Klawonn, P. D. F. (2012). Modelling Three-Dimensional
Objects. In Introduction to Computer Graphics (pp.
129–160). Springer London.
Kolbe, T. H., Gröger, G., & Plümer, L. (2005). CityGML:
Interoperable Access to 3D City Models. In P. D. P.
van Oosterom, D. S. Zlatanova, & E. M. Fendel
(Eds.), Geo-information for Disaster Management (pp.
883–899). Springer Berlin Heidelberg.
Library of Congress. (2011a, May 24). ESRI Shapefile.
Retrieved 1-28-2017 from www. digitalpreservation.
gov/formats/fdd/fdd000280.shtml.
Library of Congress. (2011b, August 5). ESRI Arc
Geodatabase. Retrieved 1-28-2017 from www.
digitalpreservation.gov/formats/fdd/fdd000293.shtml.
Library of Congress. (2016a, November 9). DWG
(AutoCAD Drawing) Format Family [web page].
Retrieved 3-7-2017 from www .digitalpreservation
.gov/formats/fdd/fdd000445.shtml.
Library of Congress. (2016b, December 7). DXF
(AutoCAD Drawing Interchange Format) Family,
ASCII variant [web page]. Retrieved 3-6-2017, from
http://www.digitalpreservation.gov/formats/fdd/
fdd000446.shtml.
Mayer, H., Sester, M., & Vosselman, G. (2013). Basic
Computer Vision Techniques. In J. C. McGlone (Ed.),
Manual of photogrammetry (pp. 517–583). Bethesda,
Md.: ASPRS.
McGraw-Hill Construction. (2012). The Business Value of
BIM in North America: Multi-Year Trend Analysis and
User Ratings (2007-2012) (SmartMarket Report).
McHenry, K., & Bajcsy, P. (2008). An Overview of 3D
Data Content, File Formats and Viewers (No. isda08-
002) (p. 21). National Center for Supercomputing
Applications: University of Illinois at Urbana-
Champaign, Urbana, IL.
Molenaar, M. (1989). Single valued vector maps-A
concept in Geographic Information Systems. Geo-
Informations-Systeme, 2, 18–26.
Muller, P., Cohn, J., & Nicholson, D. (2004). Immersing
Humans in Virtual Environments: Where’s the
Holodeck? Presented at the Interservice/Industry
Training, Simulation, and Education Conference
(I/ITSEC) 2004.
Nagel, C., Becker, T., Kaden, R., Li, K., Lee, J., & Kolbe,
T. (2010). Requirements and space-event modeling for
indoor navigation. Tech. Rep. OGC 10-191r1, Open
Geospatial Consortium Discussion Paper.
OGC. (2010). The OGC Abstract Specification, Topic 2:
Spatial referencing by coordinates (No. OGC 08-
015r2).
OGC. (2012, April 4). OGC City Geography Markup
Language (CityGML) Encoding Standard.
PDF3D. (2015, July 1). About PRC Format. Retrieved 1-
28-2017 from www.pdf3d.com/wp-content/cache/
supercache/www.pdf3d.com/about-prc/index.html.gz.
Petrie, R. (2016, March). buildingSMART Data Dictionary
Strategic Overview. Retrieved from buildingsmart.
org/wp-content/uploads/2016/03/16-03-21-bSDD-
Strategic-Overview-Webinar.pps.
Popescu, V., & Mudure, M. (2008). Interactive
Photorealistic Inside-Looking-Out Automated 3-D
Modeling. In Interservice/Industry Training,
Simulation, and Education Conference (I/ITSEC).
Reinhardt, J., & Bedrick, J. (Eds.). (2016, October 19).
2016 Level of Development Specification. BIM
Forum. Retrieved from bimforum.org/wp-
content/uploads/2016/10/LOD_Spec_2016_Part_I_20
16-10-19.zip.
Rich, S., & Smith, J. (2014, July 11). Introducing the
Facilities Information Spatial Data Model at the 2014
Esri User Conference. Retrieved 1-28-2017 from
www.prweb.com/releases/2014/07/prweb12009858.ht
m.
Safe Software. (2017). All Applications & Formats
Supported by FME. Retrieved 3-6-2017, from
www.safe.com/fme/formats-matrix/#!
Shephard, N. (2015). Ten Questions for Esri: 3D Expert,
Nathan Shepard. Retrieved from geonet.esri.com/
thread/168231-ten-questions-for-esri-3d-expert-
nathan-shepard.
Spicer, R., McAlinden, R., & Conover, D. (2016).
Producing Usable Simulation Terrain Data from UAS-
Collected Imagery. Presented at the
Interservice/Industry Training, Simulation, and
Education Conference (I/ITSEC) 2016.
Stephens, R. (2008). Semantic Object Models. In
Beginning Database Design Solutions. Wrox.
Van Sickle, J. (2010). Basic GIS coordinates (2nd ed).
Boca Raton, FL: CRC Press.
Web3D Consortium. (2008). Extensible 3D (X3D),
ISO/IEC 19775-1:2008. Retrieved from
http://www.web3d.org/documents/specifications/1977
5-1/V3.2/Part01/components/group.html.
Web3D Consortium. (2017). What is X3D Graphics.
Retrieved 1-28-2017 from www.web3d.org/what-x3d-
graphics.
Zlatanova, S., Sithole, G., Nakagawa, M., & Zhu, Q.
(2013). Problems in indoor mapping and modelling. In
The International Archives of the Photogrammetry,
Remote Sensing and Spatial Information Sciences,
Volume XL-4/W4, 2013. Cape Town: ISPRS.
Zlatanova, S., Stoter, J., & Isikdag, U. (2012). Standards
for Exchange and Storage of 3D Information:
Challenges and Opportunities for Emergency
Response. Proceedings of the 4th Int. Conference on
Cartography & GIS, Volume 2, Albena, June 2012,
17-28.
Modeling Standards and File Formats for Indoor Mapping
275