And the Oscar for geographic information systems goes to…. When the film, “Blade Runner 2049,” received the Academy Award for Best Visual Effects, the geospatial community earned some bragging rights. Blade Runner’s future city was created, in part, using GIS and geospatial tools.
Framestore, which provided the Oscar-winning visual artistry behind the special effects in “Blade Runner 2049,” prides itself on its use of new technologies. In the case of Blade Runner, this included CityEngine, the 3D modeling software designed by Esri. How did a tool designed for the geospatial profession to support city planners and architects become part of an Oscar-winning film?
“We were tasked in creating a futuristic Las Vegas for the year 2049. We had to fly over the main strip of Vegas on a long sequence shot,” says Didier Muanza, lead environment TD for Framestore. “CityEngine seemed to be the best choice to help us generate a modern-day model of Vegas, which was used as the base for the futuristic version.”
Esri, when it announced the Oscar nomination, explained, “CityEngine is a sophisticated 3D modeling software used in professions such as urban planning and architecture as well as film and other entertainment industries. Framestore designers used CityEngine for only one month and generated a set that was 6.759 km long and 4 km wide and included more than 1,400 buildings. Realistic textures were based on aerial and oblique photos of the Las Vegas main strip.”
Layers of Complexity
In an interview with the BBC program “Click,” Richard Hoover, visual effects supervisor for Framestore, described some of the process of developing the future landscape. He explained that one of the demands was to create a future view that was based on what we know of Las Vegas now. They started with government mapping data covering the Vegas Valley and city of Las Vegas.
Dennis Gassner, the art director of the film, built a simple model of Las Vegas with “buildings” designed by Syd Mead, Hoover continues. The next step was to bring in human elements and scale. Working from the paintings of the future buildings, Framestore designers placed the futuristic items around the city in an organized way to make it look like people were there. Even without people, this effort at realistic visualization provided a sense that people were living there.
Hoover describes the creation of the “Trash Mesa,” saying it was based on the idea that everyone had moved to the city and the structures outside of it were abandoned. The trash from the city was shown being dumped into the landscape, including the abandoned buildings. To model the landscape, they took a portion of present-day southern California – from Los Angeles to San Diego. They added photographs of the groundscape from Iceland and layered one on top of the other to form the future countryside.
Another view is based on the actual ship harvesting operations in Bangladesh where scrapped vessels are being dismantled.
Hoover notes the result came from grabbing all of the various components that were based on today’s reality and scaling them. He told BBC, “It was a matter of pulling off that scale and that distance, which was a massive undertaking just on the amount of data and assets we had to build and things we had to manage….”
Back to Reality
The dystopian landscape of “Blade Runner 2049” may not be one we want to build for anything other than entertainment, but the film does help demonstrate the power of the tools available to urban planners and architects who want to project the possible impacts of growth and development or show a proposed construction project in context.
As with the film, the planning needs to start with a current view. POB asked Esri to put things into a real-world context and show what is necessary to build that foundation, and then evolve images of a potential future city or community that would be far more desirable and livable than that of “Blade Runner 2049.” Here are some comments from Taisha Waeny, project manager at the Esri R&D Center in Zurich, Switzerland.
POB: How should cities plan for the survey and GIS data collection?
ESRI: Nowadays, we have so many different methods of capturing data, all of which give us great accuracy (mostly) and are hugely beneficial when it comes to creating a solid foundation to build a good GIS. On the one hand, we have the primary data collection sources like remote sensing, traditional surveying, GPS and LiDAR.
LiDAR is certainly a technology that is favored these days because of the fact that you can get really detailed datasets quite quickly (especially in comparison to more traditional surveying methods).
There are different sources of secondary data capture available like aerial imagery and photogrammetry from which you can get additional raster or vector data. It’s even possible to get citizen-based data these days. A more advanced city would employ all of these techniques to set themselves up with a solid GIS.
POB: You mention citizen-based data, which may not maintain the same levels of control as other sources. How can the different levels of data be blended in the resulting model to provide the most usable result?
ESRI: With regard to citizen-based data, 100 percent as-built scans are more desirable than citizen-based for anything construction related. I mentioned citizen-based merely to highlight that there are more and more ways of gathering data these days. Citizen-based data would probably be used more in a context of nature preservation, not necessarily for construction. Let’s say that there are plans to build on a paddock. The way the construction company could go about it would be different if this was a place where lots of rare flora and fauna have been found and sighted. The different data collection methods are more like augmentations to the standard data that is needed for building and construction.
POB: Is there a clear distinction between what data needs to be survey grade and what can be GIS grade?
ESRI: I think that these two points really go hand in hand. As mentioned, a lot of surveying methods are used to actually derive GIS data. It really would depend on the use-case as to which accuracy the data would need to have and which methods are employed to achieve it.
POB: With the data collected, what tools are used in a current smart city scenario? And what is the result (asset management, infrastructure planning, etc.)?
ESRI: I think, again, this is really use-case dependent. A smart city can encompass so much. Just to name a few examples, for urban planning and simulations, the user might use Esri CityEngine. Esri CityEngine is the desktop application that Framestore used to create the city of future Las Vegas. Using CityEngine, large-scale urban environments can be created procedurally – this means using code to enter in all kinds of parameters that might be useful like gross floor area, building usage, calculating visibility, etc.
We also need to take into account that the whole building and construction industry is becoming more digital and that interoperability and accessibility are increasingly vital. This is one of the reasons why Esri has also recently partnered with Autodesk. This partnership is bridging the gap between GIS and BIM. Utilizing these tools, having access to all this information, and being able to understand demand, capacity, suitability, allocation, etc., obviously saves a lot of time and allows for a much better return on investment.
POB: Where can the city take the planning and modeling once this basic mapping tool has been developed?
ESRI: I think I would almost go as far as saying that we are talking about more than just one tool here. I think of GIS more like a system that leverages the capability of many tools. Within the system, we would have specific tools for different needs. I would say that the core value of GIS is that it adds efficiency.
If a city is using GIS and data capturing technologies well, then we can start living in a more predictive environment. As an example, cities would know where all their sewer lines are and what kind of state they are in, when they require maintenance, what kind of pipes they are made of, what the volume of the pipes is, etc.
We also already live in a world where artificial intelligence can be used to analyze and model data, so I really do believe that as we go on, this GIS ecosystem will continue to grow to levels even beyond what most of us are capable of envisioning.
POB: You mention the sewer line example, where city planners and maintenance can essentially do a virtual fly through to evaluate their response to infrastructure needs. The tools and data related to infrastructure may be in many different hands and the availability may not be apparent. What is needed to be able to locate and incorporate data from these disparate sources?
ESRI: The beauty of data and GIS is that all data can be put into layers, and with the use of all the Esri tools for example, you can create a web-based GIS system that would allow for anybody who needs/wants it to access it. If you have editing privileges, you can edit the data (e.g. update the as-built scan) and the most up-to-date version will be immediately available to everyone.
The same concept goes for the different parties required in construction: the surveyor uploads the most up-to-date as-built/stake out survey results. The foreman immediately has access to the newest data and plans. The architects change their plans. The finance guys know about it straight away, as do the builders. This immensely reduces errors that are made due to the lack of interoperability. Everybody is always working from the same page. No more nasty surprises.
POB: I’m assuming some degree of interoperability here so that the issue is finding the data, not integrating it.
ESRI: I think this also answers your question. The infrastructure is there to ensure interoperability. I think it is up to the councils to create a platform that would allow for easier collaboration between various parties (if we’re talking about this on a city-wide level). We already have some publicly available data in the form of Open Street Map data, but this concept can be taken further too.
POB: Even though the futuristic Blade Runner set represents a fictional future Las Vegas, can you put some of that model into a realistic context?
ESRI: Hmmm…good question. In a film, it’s obviously always important to show exactly what can be seen from the camera, and first and foremost in movies, it’s all about great visualizations and quality renderings. In the real world, we need to understand the actual terrain, so a lot of care needs to be taken that the environment is completely modeled. To get a fully fledged 3D model of, in this case Las Vegas, it would require a whole lot of effort to capture all the data. Due to this, I guess some effort is spared when it comes to the visualization of this data. So, although so much work has gone into this, it may not be quite as appreciated as a beautifully rendered scene in Hollywood – realistic detail is given up for a more realistic data set.
But the beauty of GIS is that this is a system where all the data can be aggregated and visualized both in 2D and 3D, even if it’s not quite Hollywood quality.
I digress a little. To answer your question more eloquently, I’m not entirely sure which exact data sets Framestore used when developing the city of “future Las Vegas,” but depending on the data used in “Blade Runner 2049,” perhaps some assumptions can be made for the future.
In a Theater Near You?
Framestore used some of the datasets and tools that are common to GIS as a foundation to create a fictional future landscape in “Blade Runner 2049.” They wove in some visual artistry for a Hollywood-grade rendering that netted its team an Oscar. Urban planners, architects, engineers, and construction contractors can apply many of those same tools to a present-day setting and project a finished view of the future they want to help build. Though it may be a little less visually stunning than the Hollywood-grade images, the 2D and 3D renderings can be powerful tools for creating that future. As with “Blade Runner 2049,” it all starts with a map.