Introduction
The recent technology of virtual globe-based 3D visualization is a unique opportunity to facilitate advanced
analysis and visualization tasks in a variety of application domains such as urban planning, indoor/outdoor
pedestrian navigation, environmental simulations, cultural heritage, or facility management. One core data
which make this concept come true is the development of the CityGML models by Open Geospatial Consortium
(OGC — https://www.ogc.org/). CityGML is the global data model schema for describing 3D
geo-spatially-enabled urban models. There are more and more cities that provide open-source CityGML models
publicly.
(An Expected Output from This Workshop — 3D Web App with 3D Semantic Building Models in Den Haag Area)
Architecture and Workflow
The overall system architecture of the 3D web-based app is shown in the figure above. It includes four main
components: CityGML dataset, 3D Tiles dataset, Web server, and Users.
So, let’s get started and create each component together
Step 1: Preparing 3D City Models
In this step, we will download the 3D city models in CityGML format. Then, convert the models to 3D Tiles
format for optimizing the 3D web visualization. In this article, I picked the 3D city models from Den Haag
area as it is a compact model. Of course, you are free to explore this method with any CityGML datasets.
Download the CityGML dataset from
Using the FME Workbench to convert the CityGML models to 3D Tiles format. Starting by opening the FME
Workbench, Drag and drop all CityGML models to the FME window.
Select input Format "CityGML" and Workflow option "Single Merged Feature Type".
Click on "Add a new writer" button and select Format "Cesium 3D Tiles"
The workflow should look like following figure. Click Run and then wait until the conversion complete. Then,
the final 3D Tile City Models will be in the destination folder you input.
Step 2: Preparing HTML for CesiumJS
In this step, we will create an HTML web document containing the 3D Web Globe using CesiumJS library. You can
use any code Idle or editor. You can create the document with
the name you like and copy the following code. Don't forget to specify the URL on line 24 to the target
Tileset.json. If you do this for the first time, you can come back after you set up the web server in step
three.
Step 3: Host the 3D Web App on the Web Server
This steps, we will host the Web App from step 2 and the 3D Tiles city models from step 1 on the web server. You
are freely to use any web server of your preferences.
3A: Python web server
Install Python (3.xx)
open the CMD (Command Prompt) and check the python version python --version. If you get an error
message, please check your python installation.
Start the server by python -m http.server 80
3B: XAMPP (Apache) web server.
(You can get it from
https://www.apachefriends.org/index.html)
Install XAMPP or Apache.
Open XAMPP Control Panel and start Apache HTTP web service.
Optionally, you can start the Apache
manually via [(part to Apache)]\apache\bin\httpd.exe
. Or, use Terminal to enter sudo
apachectl start
and sudo apachectl stop
commands to start and stop Apache service.
Place the web document (denHaagApp.html) from step 2 and a folder of 3D Tiles city models from step 1 in the
C:\xampp\htdocs directory.
In the html web document, update the URL value to the tileset.json file of your 3D Tiles model in the htdocs
folder. For example, if your 3D Tiles directory is at C:\xampp\htdocs\3dtile\tileset.json. You should set
the url value to "./3dtile/tileset.json" or "http://localhost/3dtile/tileset.json" as APACHE will serve
everything in the htdocs directory relatively to "http://localhost".
Step 4: Test the 3D Web App
Finally, you can access your 3D web app visualizing 3D city model on your web browser.
Overview of the software used
Other Example Use Cases
Deutschland in 3D 3D Web Application Visualizing
German Cities
OGC 3D IoT Pilot - Sejong, South Korea with air
quality sensor data (SensorThings API)
3D Visualization Server for hosting 3D building models
created by the 3D GeoVolumes API (3D Container API)Specification
Explore other CityGML Datasets
Copyright
The texts and images provided in the blog are under the CC BY 3.0 DE license.
Author
Thunyathep Santhanavanich (JOE) PhD candidate
Faculty of Geomatics, Computer Science and Mathematics,
University of Applied Sciences Stuttgart,
Schellingstr. 24, D-70174 Stuttgart
thunyathep.santhanavanich@hft-stuttgart.de