Create your own 3D model with Open Data!
Creating a 3D model can seem like a massive project. First you need a building footprint layer and then have an aerial assessment completed for your area of interest. You also need a decent knowledge of 3D geoprocessing tools and must minimize the size of the data so that it can be published in ArcGIS Online or in ArcGIS Enterprise portal. In this blog post, we will show you that the process is not as difficult as you think and for most municipalities in Québec, the data is at your disposal and at no cost.
You've surely seen 3D models hosted in some of our Esri products. ArcGIS Urban, for example, is a product that helps transform urban design and planning in smart cities to address multiple issues like housing availability, sustainability goals and economic fluctuations that force cities to better plan for the future (view an interactive demo here).
At the same time, ArcGIS CityEngine allows conceptual modeling and creation of 3D buildings and cities in a fast and efficient way using the procedural rules app (view an interactive demo here).
You must have also heard of integration of BIM (Building Information Modeling) with GIS in order to add contextual information such as flood zones and geology throughout a construction project (view an interactive demo here). This integration is part of the “feuille de route gouvernementale pour la modélisation des données du bâtiment (2021-2026)” (government roadmap for building information modeling (2021-2026)) (in French only) published by the Government of Québec.
Our goal is not to present every single 3D solution. We simply aim to show that if you have an ArcGIS Pro Advanced license with 3D Analyst and Spatial Analyst extensions, you are already able to create your own 3D model in a short period of time.
In the coming sections, we will start by introducing ArcGIS Solutions gallery, representing the most commonly used workflows in GIS. We will then show you how to acquire the data needed to create your own 3D model and finally we will demonstrate how to customize the texture of your buildings.
For demo purposes, we have created two 3D models. The first model represents the City of Québec’s InnoVitam project (in French only). This project is focused on preventive and sustainable health bringing together commerce, residential, business and green spaces (see the 3D model here).
The second model represents the plan directeur de revitalisation du centre-ville de Trois‑Rivières (masterplan for revitalizing the downtown area of Trois‑Rivières) (in French only) to demonstrate the underground infrastructure in 3D (see the 3D model here).
Please note: the design for InnoVitam and for Trois-Rivières 3D models was produced entirely by Esri Canada and are not in any way affiliated with Ville de Québec nor with Ville de Trois‑Rivières. All of the data was obtained from Données Québec (in French only).
ArcGIS Solutions – 3D Basemaps
As previously mentioned, ArcGIS Solutions gallery consists of ready to use solutions employing the most commonly used workflows in GIS. The solution used to design our models is called 3D Basemaps. For more information, please refer to this blog post published by my colleague, Maggie Samson of Esri Canada.
Explore the different tasks available within the solution. You can create your 3D model from raw LiDAR data, as shown in the Extract 3D buildings from lidar data training session, or from derivative products like a digital terrain model (DTM) and a surface model (DSM) by going through the steps in the Extract 3D buildings using photogrammetry training session. For Québec and Trois‑Rivières models, we used the photogrammetry process.
Building Footprint Layer
When you are ready to deploy 3D Basemaps, you will download an ArcGIS Pro project containing several detailed tasks. One of the first tasks is to estimate the height of the buildings and the shape of the roofs. To do this, you need the surface building footprint layers for the desired area. The footprint layers were obtained from Données Québec (in French only). The data can come from several sources including municipalities, the ministère des Ressources naturelles et des Forêts (MRNF – formerly ministère des Forêts, de la Faune et des Parcs) or Ressources naturelles Canada (NRCan) (in French only).
Ground and Surface Digital Model
To determine the approximate height of the buildings and the of the roofs, we downloaded the digital surface (or canopy) and terrain models from the ministère des Ressources naturelles et des Forêts (in French only). LiDAR-derivative products are generated as part of the Projet d’acquisition de données par le capteur LiDAR à l’échelle provinciale (the province’s LiDAR Sensor Data Acquisition Project) (in French only). The derivative products have 1m of spatial resolution and consist of most of the Québec meridional (south of the 52nd parallel). The products are divided into several layers for downloading. You can access the Forêt ouverte (Open Forest) (in French only) app to download LiDAR derivative products. If you are using another source for digital models, a minimum resolution of 2m is suggested for the photogrammetry process.
Once the 3D model has been created, you can modify the symbology and the texture. When deploying the 3D Basemaps solution, a file named "rule_packages" containing several 3D symbologies is downloaded locally. Procedural rules (.rpk) are generated by ArcGIS CityEngine and are used to symbolize your buildings according to attribute values. Please note that ArcGIS CityEngine is required only if you want to create your own procedural rules.
You can find additional procedural rules in the ArcGIS Online atlas (example here). Once you are satisfied with the product, follow the publishing steps within the solution to deploy the scene in ArcGIS Online or in ArcGIS Enterprise portal.
And there you have it! You have just created your first 3D model with open data exclusively. The advantage of the methodology that we have demonstrated is of course that the data is free and that if you have ready access to extensions, you can begin to develop your models now.
If you want greater accuracy than what is shown in this blog post, you can create a model from your own LiDAR data or you can acquire the raw data from shopping sites such as MRNF’s Géoboutique (in French only). By using LiDAR point clouds, you can get the most out of 3D basemaps by creating 3D models of bridges, trees, water levels as well as overhead power lines.
Don’t forget that you can also obtain 3D data in OpenStreetMap for trees and buildings and that you already have access in ArcGIS Online, regardless of your ArcGIS license. However, the data is maintained by a group of volunteers from all over the world and there are some strong possibilities of inaccurate data in your area of interest. You are also limited in the edits you can make to the data since it does not belong to you.
No matter where you are in your 3D GIS strategy, the goal of this blog post was to show you that creating 3D building and infrastructure models can be easy to do and can be a first step in enabling your organization's transition to digital twins and to a smart city.