Visualization is only one aspect of 3D. At Esri, the term visualization means something different than what it means in terms of engineering. Usually we refer to visualization as making a map. But 3D exposes GIS vendors to new users, such as real estate, architecture or infrastructure customers, who expect higher resolution representation of realistic visuals and visual effects expressed in cartographic expression in 3D. 3D Visualization World Magazine interviewed Chris Andrews, Product Manager for 3D across the Esri ArcGIS platform to learn more about visualization and 3D at that company.
How did you become interested in 3D and visualization? What sparked your interest in your current work?
The first time that I ever used an official GIS product, which was an Esri product in the 90’s, I wanted to tilt the map. My background is in ecology and geology. During my time in geology we hand drew strata maps, often in 3D perspective. Geologists measure strike and dip to understand the direction of rock layers from the surface, so when I was in the field we were thinking in 3D. During my ecology work with ant populations, I used 3D measurements to attempt to visualize the farm fields where the populations were found.
As far as getting into GIS, I had essentially built a custom GIS to map ants during my MS work. I then taught myself basic GIS and, after graduation, I found jobs quite quickly with my programming and GIS skills.
I was introduced to Esri 3D Analyst when it first came out but it wasn’t until I worked at Autodesk that I really dove into 3D. I helped lead the Autodesk Digital Cities project for four years. This effort resulted in the product that eventually became Autodesk Infraworks.
What was the idea or impetus behind your work on the Autodesk Infraworks effort?
The idea was to provide a solution that could drive the flow of GIS and engineering information between cities and their contractors. The effort looked at data flow for construction and design supported by Building Infrastructure Modeling (BIM) as well as asset maintenance and facilities workflows. The Digital Cities effort was intrinsically conceived as a 3D effort from the start. The primary view at Autodesk was to view the project as a geospatial 3D effort. Our group was focused on cities and infrastructure. Once we started to interview architects and engineers and work with adjacent software teams in Autodesk, such as the AutoCAD Civil 3D and Autodesk Revit teams, then it was clear we had a 3D problem set.
In terms of Esri - why is 3D and visualization important?
3D is important to Esri because our customers, fundamentally, find value in it. Visualization is only one aspect of 3D. At Esri, the term visualization means something different than what it means in terms of engineering. Usually we refer to visualization as making a map. Traditional maps may have thematic colored information, expressing some analysis or output. In the architecture and engineering world, visualization is often viewed in terms of selling projects or proposals – a different context from displaying information. Visualization in the GIS world is often more cartographic in nature. Design Visualization for engineering on the other hand strives for precision and realism, when appropriate.
What have you found interesting since coming to Esri in view of these two approaches mentioned above?
The GIS world has generally seen 3D as a new extension to 2D mapping and sometimes attempted to just shift 2D cartographic concepts into 3D. But 3D exposes GIS vendors to new users, such as real estate, architecture or infrastructure customers, who expect higher resolution representation of realistic visuals and visual effects expressed in cartographic expression in 3D. The value of the ‘map’ is different to these users in 3D.
What are the key values of 3D in your mind?
3D enables users to see things that either don’t yet exist or which can’t be easily seen, such as underground information. 3D requires a higher level of accuracy to generate useable products and some are starting to see that 3D can lead to a more accurate picture of the real world. Many stakeholders who are not trained to read 2D maps find it easier to understand concepts and proposals in 3D. Finally, there are some kinds of analysis that cannot be done in 2D, such as wind flow and microclimate analysis around buildings in an urban setting.
GIS are data driven, perhaps the better question to ask is “what is the connection between data and 3D?”
Data is one of the biggest enablers of any kind of GIS work, perhaps even more so in 3D than in 2D. Esri has worked hard to enable customers who have 2D data to be able to use that data in 3D. For example, customers can use procedural technology to extrude realistic buildings from 2D building footprints or they can create spatial analyses that allow scientists to see time variation by displaying 2D information in 3D space.
Many users are now collecting realistic 3D information about the environment around them, capturing point clouds or phototextured meshes which can generate massive data sets. For these users, we have proposed new standards and specifications to allow sharing and distribution of large amounts of 3D content for visualization and analysis. We released the Indexed 3D Scene (I3S) layer specification so that partners could more easily create 3D content that can be consumed across Esri apps. Now I3S is working its way through the Community Standard process in the OGC. Vricon is a good example of a partner who is now using I3S to deliver large volumes of data to ArcGIS users.
3D has many challenges. Could you tell me about the challenges you see in terms of your own organization and trying to build capacity there as well as outside of Esri?
The GIS community has traditionally oriented around 2D. In fact, there were arguments and questions about why one would ever want or need to use 3D. In the past, you saw, for example, articles stating that true 3D is too high density to do analysis. Now we have an emerging field of HD Maps that is intrinsically 3D and used for high accuracy, real-time analysis.
The fact is, that requirements for 3D and 2D depend on what kind of problems that need to be solved. While one might capture 1 mm data for micro-drainage analysis for a specific roadway, data of that accuracy is not necessarily needed for studying the impacts of sea level rise in Miami. Getting people to understand that some kinds of problems require 3D data has been one challenge.
Within Esri, I promote the idea that the focus is neither 2D nor 3D but that we should focus on advanced GIS circa 2025 and beyond. The products we are building today will be competing in the marketplace in 10 or 15 years. ArcMap, as an example, was originally released around 2000 and is still important today. We need to understand the future with respect to the trajectory that our products are setting today in the context of rapidly emerging and changing technologies.
In every early market, there is also a dichotomy in customers’ behavior and expectations that slows down adoption. Customers will try a consumer-type AR or VR experience, for example, and say they want that experience but with their GIS data. To even get them started on the early capabilities to make this happen, the customers need to migrate to newer versions of technology, which they resist at the risk of changing existing workflows. In an early stage market, there are few examples in production where new technologies add real value, in this case 3D with AR and VR which are still being proved out. It can be difficult to get customers past this hurdle and into upgrading to a higher level of capability to help them explore the future. Customers need to see value and we have to help them demonstrate that value.
It seems that both hardware and software sides are constantly changing in the 3D space. What issues in terms of innovation rate in 3D do you find challenging at Esri?
Yes, there are both new data types and old data types that are suddenly more popular. Spherical-imagery is an example of an old data type that is suddenly popular. Its recent proliferation beyond architecture or hobby use has come up very quickly. Cheap consumer technologies and industrial strength capture technologies are collecting spherical imagery regularly indoors and outdoors with geolocation sometimes in the millions of images at a time. People want to use this data immediately with their GIS.
Drone-based data collection is another example. About 2 million drones were sold last year in the US. Many of these drones are being used to generating huge amounts of imagery at a low cost that can then be used to generate point clouds and 3D meshes. The diversity of data types in use today and the sudden availability of massive data sets is a big challenge for the industry.
All of these questions seem to point to the fact that we do not have enough people who understand 3D enough to build and create the applications we want? Is that a fair statement?
We are constantly on the lookout for qualified people to help us implement the future of 3D and GIS. Jack Dangermond has told me many times that if we can find the people, I should go ahead and hire them. We can also see partners who have been slow to get more involved in 3D. Customers are leading 3D requirements for new projects which tends to push partners forward. I’m finding many customers who have a vision that includes 3D and GIS. Universities, such as the University of Kentucky, are now embracing BIM and the BIM world well established the utility of 3D in the approval process for new buildings before they are built.
We used to have these long, great battles between GIS and CAD domains. Do you think we are finally over all that turmoil?
I don’t hear that much mention of the war between GIS and CAD these days. The traditionally CAD-oriented industries are far more focused on BIM, which is a process more than a technology.
Cities in their lifecycle spawn capital projects. Every capital project, should be conducted with a BIM process because BIM has been shown to save vast amounts of money and time during construction. Some groups conflate cities’ lifecycle management processes with BIM. Most cities do not do construction, they plan, provide services, and collect taxes with the help of GIS. Cities outsource construction to providers who will use BIM.
In my opinion, some academic groups have placed unnatural expectations on BIM which confuses users. I’ve seen articles or discussions that imply that GIS can do BIM and BIM can do GIS. The reality is, GIS and BIM are used to solve different problems in the asset lifecycle.
Does 3D and visualization bring people in to understand GIS and spatial data? Where is Esri now?
Definitely. It’s far easier for a typical person to understand viewshed restrictions, right-to-light regulations, and subsurface cadaster in 3D with GIS data, for example. To enable more users to benefit from the use of 3D, we need to be able to supply every ArcGIS experience with 3D content and enable users to easily create compelling 3D visualizations. We are actively working on 3D in just about every part of the ArcGIS platform.
What is the connection of 3D and GeoDesign in terms of Esri – it does not seem clear sometimes?
GeoDesign is often equated with planning, but that term is too general. I think of GeoDesign as using GIS and other information in a community-based process to influence regulatory guidelines or legislative outcomes that drive a successful community. 3D is an important part of GeoDesign because 3D can be used to provide accurate analyses and because 3D helps communicate plans and proposals.
Sometimes GeoDesign is confused with the engineering preliminary design process in which rough engineering designs will be shown in virtual location context, often for approval of a specific initial design. Typically, the outcome of the engineering preliminary design process leads to rough models that will be used in the next stage in the detailed design phase, leading up to construction. The preliminary design process also happens in 3D and can use GIS content, but is unlikely to be done in GIS software.
Where does Esri stand on modeling and simulation?
I’m starting to hear more modeling and simulation expectations from GIS customers as speed and performance improve for standard GIS software, but I also think that modern technologies are changing the modeling and sim industry. We can see emerging modern computing architectures that are starting to approach with live GIS content what a traditional modeling and sim approach might have achieved with a ton of data preprocessing, for example. The urban data analysis and Internet of Things (IoT) markets are likely going to drive GIS technology to process data and produce visualizations that look a lot like some of the work that has been done in the modeling and sim space.
Chris Andrews is the Product Manager for 3D across the Esri ArcGIS platform. During his 20-year career, Chris has focused on exposing the value of GIS throughout organizations by combining cutting edge technology with ease of use to put the most capability in the user’s hand with the least amount of specialized technology expertise required. Early in his career, Chris taught the first GIS classes at the University of Rochester and then began a software career by joining startups to work on digital mapping products for law enforcement and real estate professionals. An interest in web technology and multi-tier systems took Chris to the Kennedy Space Center for a Center-wide GIS requirements and architecture project which launched him into Enterprise consulting where he worked with law enforcement, local government, telecommunications, utilities, and commercial customers.
Chris’ interest in 3D began in college, but took off when he joined Autodesk in 2007. As the first technical product manager for Autodesk’s 3D digital cities effort, Chris led pilot studies with three globally distributed cities, Vancouver, Salzburg, and Incheon, to gather requirements and evaluate technology. After taking over as the lead product manager on the effort, Chris led the development of a 3D engineering preliminary design product that is now widely used in the AEC industry. Joining Esri in 2014 allowed Chris to begin exploring the use of 3D across multiple domains including defense, intelligence, commercial industries, and the sciences. At Esri, Chris remains focused on distributing high value GIS more broadly within organizations, initiated the ArcGIS Earth product effort, and is now focused on BIM-GIS integration.
Chris holds a Bachelor of Science in Geology-Biology and a Master of Science in Ecology and Evolutionary Biology from the University of Rochester where he concentrated on spatial clustering analysis and spatial population simulation modeling.
For more information: 3D at Esri