Jumpstart Your Project: Cutting Edge Imagery Workflows with AI and 3D

STEVEN SANTOVASI: Welcome to our presentation, "Jumpstart Your Project-- Cutting Edge Imagery Workflows with AI and 3D." My name is Steve Santovasi. I'm here with Jeff Saunders. We're going to present this to you today and hope we can teach you a few things. I'm going to start here with a safe harbor statement that we may be showing some forward-looking statements or future developments and efforts for our new or existing products or services.

And just know that purchasing and investment decisions should not be made based upon reliance of any of the statements that you see here today. All right, thank you. So we'll get started here with the agenda, what to expect from this session. So essentially, we're going to start with introductions.

We're going to go through a little bit of the anatomy of a project, what's in a project, the teams, the people, the timeline. We're going to talk about great imagery content starts with the camera system. So we'll talk about the capture systems. And then, we're going to get into data. What's in a picture?

The high quality imagery products that are created from this process. And then, I'm going to turn it over to Jeff. And he's going to talk about applying those imagery products to Autodesk design workflows. And at the end, we're going to round it out with a big picture. And then, we're going to go to some Q&A.

So what's in a project? It takes many teams to successfully complete a large-scale infrastructure project. Inside of any project, we know that there's different phases-- in planning, design, building and construction, and operations and maintenance. We'll take a look here at who are those project stakeholders?

Who are all the people that are involved to help make this a successful project? Could be anything from your data engineers, your professional engineers, your project managers, your asset managers, your BIM technicians, your CAD drafters. There's all kinds of people who are involved to make this happen. And at every step of the way throughout this process, each one of these teams has deliverables that need to go and support a certain phase of the project.

So let's take a look at the anatomy of an AEC project lifecycle. So we think about an asset. And we say, it needs to start with a master plan. Do we need to build this project? Is this something that we need? In order to understand that, we have to go through analysis. We have to go through site selection, demographics.

And ultimately, we end up with a master plan report. And once we have that master plan report, we can then decide, should we build this? And once we decide that we're going to build this asset regardless of what it is or how large the project is-- it could be a $100,000 project. It could be $100 billion project. We have different pieces that we go through throughout this lifecycle. So the next step is the regulatory compliance.

So we have to go through our permitting process, whether it's building permits, zoning, wetlands, public hearings. And at the same time, we start to understand our stakeholder management, our public outreach, our real estate, our access rights so that the public is involved and they understand what this project is going to look like.

At the same time, we start our preliminary design. We go through our reality capture, our constructability reviews, our topo surveys. We look at environmental threatened and endangered species and cultural studies to be able to say, yes, we all agree this is where we should be building this and we know where we shouldn't be infringing on other areas. That leads us into the detailed design phase of this, where we actually start our civil design or our BIM design.

We start to put together our schedule and our document strategies. And then, throughout all of this, we have safety and we have environmental monitoring. And those are two of the most important pieces that we have here. And once we have all this set up, we start into our pre-construction, where we look at our bid packages. We go through our procurement process.

We prepare our site. We start doing our staging. And then, we get into the actual physical construction, where we have project control teams, project management teams, construction managers. We go through QA/QC processes. We look at our daily reports. And at the end of that construction, we have an asset. We need to restore and remediate any damage that we may have done to the environment surrounding that the project site.

And then once it's constructed, we then have another phase of turning over all of that documentation and there could be, in some cases, millions of documents-- CAD files, GIS files, BIM files, databases, PDFs, documents, and all that needs to be turned over so that they can do audits and financial reviews and then begin the operations and maintenance phase, where once all that data is turned over, in some cases, they can build their operational digital twins where they can then connect network sensors and IoT sensors, look at the-- incorporate the [INAUDIBLE] built in it, identify what the staffing needs are to actually operate this piece of infrastructure.

And that gets us into the longest piece, the revenue and value, which hopefully, with the piece of infrastructure that we just built, we could be looking at using that for 25, 50, 75, 100 years of revenue and value. But at the end of that, we need to make a determination on decommissioning, recommissioning, restoring, rebuilding. And that brings us right back to the beginning again, to our master plan.

So if we look at this, it's a circular process that takes place here. And we look at this from the government and owner-operator side, so whether it's the government or utility and then the AEC and owners rep and construction firm that helps make that project come to life.

So with that, we understand the project phases. And we understand that they take place relatively in sequence, but there's a lot of overlap that takes place there. And what we're going to talk about today is how the imagery products can come together and help support each one of these project phases as we move forward.

So if we look at this on a timeline, we can get a better understanding of the project phase overlap-- and then, looking across this, how Nearmap imagery and derivative products can help support throughout this.

So with that being said, I'm going to move on to the Capture system. So what we've done at Nearmap is we've introduced our third-generation camera systems. And we process for results. So we look at this from a standpoint of serving our customers and serving the industry. We're dedicated to innovation. We look to raise the bar for the quality of imagery offerings.

We look to proactively capture our program areas. And that gives access to current and historic site conditions for our current customers and our future customers. The frequency of our captures helps keep that data relevant throughout any project. And the improved efficiency and enhanced processing means that we get to deliver faster. So after we capture, the idea is to get that data-- the 2D data, the 3D data, the AI data-- in the hands of our customers as quickly as possible.

So this brings value to our customers. So using these hyper cameras, we have improved 3D, especially in the central business districts. And if you look at this image to the right, it's actually a 3D mesh. It's not it's not an aerial photo. It's not a picture. It's actually a still from a 3D mesh in a central business district. So you can see the quality that we're talking about with that 3D data.

And the foundation of all of this is that we get clearer images at the same resolution. So a lot of people talk about the different resolutions at which you capture. But if you don't have extremely clear images, the ground resolution isn't nearly as important. And you're going to see here what that difference makes. We've also added near-infrared to enable vegetation analysis and then continuous expansion of our AI feature captures.

So right now, we capture over 100 different feature extractions from the imagery and create geospatial data from that. We'll do some deep dives into that in a minute.

Other products are improved digital surface model, smoother digital terrain models. And we do this all to better serve the increasingly sophisticated customers that are using our products to build digital twins. So like I said, clearer imagery is better data.

Let's look at the products that we're talking about. So everyone knows with aerial imagery, you have 2D vertical imagery. What we do is, we provide sub-3-inch imagery. And with these premium systems, in many cases, we can get less than 1 inch. We have 2D panoramic images. We offer four directions and have premium level with even higher zoom levels.

So that gives the ability to look around a building or to look around a site to be able to get more insight from that-- and, of course, the oblique imagery, where we could measure inside those obliques in 3D with a premium level of to eight different directions and multiple images per direction.

And then, we're going to talk a lot about the 3D and the AI and true ortho products. These are all derived from the HC3 imagery. And they offer significant improvements for our customers. And they drive revenue. And then, like I said, we have four-band near-infrared imagery. And we're going to talk deeper about that, about being able to utilize that infrared range to be able to create vegetation indexes, burn indexes, chlorophyll indexes, and other things.

So let's talk about the capabilities. So the opportunity that our HC3 imagery improves the Nearmap products, it provides better context and better content for design workflows. It improves the customer experience. It improves solutions for both Nearmap and Autodesk customers.

And the value that this brings to the customers is deeper insights to their built environment, improved visual content for their area of interest, improved locational intelligence, streamlined design processes, and the bottom line to help them make better decisions. So let's move on to the actual products. We'll get a quick sneak peek at what we're talking about, here-- so high quality data sets created from aerial imagery to help jumpstart your project.

I'm going to start with this first video, here. It's a quick look. I don't even have to say anything I named this, "Worth a Thousand Words." But we're actually looking at is I'm zooming in on an image. It's a tarmac.

We can see some pavement damage there. What I'm doing is continue to Zoom in to give you a feel of the level of the quality of the resolution, where right there, you can look actually inside the cracks, inside of the pavement damage there. So you can see that the quality is absolutely incredible.

So not only can we look at the image to identify things like pavement damage, we can use the AI, the mature AI models, to give us deeper insight. And like I said, right now, we offer over 120 different feature types extracted from Nearmap imagery. On the left side here, you can see, in purple, I have building footprints in green. There's different levels of vegetation.

In yellow there, I have the hard surfaces to be able to identify impermeable surface areas. And if we look at the right side, what I did there is I tried to show, number one, the raster detection of pavement damage.

So this is what the AI deep learning models are-- how they're able to identify, with a certain level of confidence, where those cracks in the pavements are and then underneath that, to be able to create vector boundaries of that data for geospatial integration into other platforms, like multiple products in the Autodesk environment.

This is a look at the city-scale 3D mesh that's being built. So if you can imagine, for your public hearings or your stakeholder management or even your tale boards, to be able to show your designs inside a full 3D mesh of the entire environment, whether it's in urban development or whether it's in a utility corridor, to be able to look in 3D at what the actual built environment looks like.

On the right side, there, that's actually another image of the 3D mesh. So that's the crisp, the clarity that we're looking at, here.

So if we start to talk about our surfaces, being able to create the digital surface model to improve what you're bringing into your design environment. So on the left side here, we can see the digital surface model draped over the 3D mesh colorized by elevation. So that's a 3D look at it. On the right side is a 2D look at a digital surface model.

That's bringing the raster in in 3D and showing those cells the elevation by color. And you could see how crisp and clear it is, the difference between the surface elevation and the tops of the buildings, there.

And then, if we look at the digital terrain model, to look at the smoothness of the actual terrain-- on the left side, I'm showing the digital terrain model, giving you a view of the surface as it is on an exit ramp and on the right side, the raster digital terrain model showing the terrain change going all the way down to the riverbed there. So you could see how nice and smooth the actual data is there.

And then, of course, as part of the process, we create full 3D point clouds to go along with it. And you can get a look here. What I did there is I overlaid a rectified point cloud over the top of an imagery there and then colored it by elevation. On the bottom-right side, you could see a point cloud there, but using the RGB value so you can look at it in full, real-life color there.

And at the top, showing an entire project work site by the point cloud, showing it by elevation there-- so to talk a little bit more about the near-infrared, we're finding more and more that our customers in government, in AEC, and in utilities have a really big use for near-infrared data.

So we've now added near-infrared sensor onto the aircraft to capture four-band imagery. From those four bands, the red, green, blue, and near-infrared wavelength, we can run different indices on it and come up with a different index, depending on the combination of the four bands that you use.

So looking at the use cases for this, we can start to understand vegetation health to help us with urban plannings for parks and campuses in green cities to monitor vegetation growth, look at growth patterns in how the vegetation is growing or not growing over time, to help with vegetation management programs for utility corridors and access roads.

In the agricultural side, of course, crop yield, water, and fertilization are all visible through near-infrared. And then, in other areas, we can help with fire risk assessment to help government agencies, utilities, fire departments, emergency management crews, and even FEMA through the use of this near-infrared imagery.

So if you look at the bottom-left there, that's a look at the near-infrared image. And on the right side, I ran the indices on four different indices-- the normalized vegetation, the NDVI, the false red the color infrared there, chlorophyll index, and then a burn index.

So you can see there's multiple options for extracting value from near-infrared imagery. And then, I want to talk about just two more topics before I hand it over to Jeff. One of them is inside of Nearmap's visual platform, which is called Map Browser, we now have the ability through what's called Geodata Link, which is our connector for ArcGIS hosted data.

So now, inside of Map Browser through Geodata Link, it enables you to connect to ArcGIS services directly in your map browser. So what that does is it allows you, when you have your design data that's created in the Autodesk environment, whether it's in civil 3D or map 3D or InfraWorks, to be able to publish that data out to an ArcGIS service and then have that design data brought back into the Nearmap Map Browser environment.

That can help with stakeholder management encroachments, constructability walkthroughs, desktop inspections, things like that. It gives us the ability to leverage the Autodesk and Esri partnership. It gives you the ability to leverage your GIS investment. And then, we have a whole bunch of things that we can do to be able to set custom zoom levels, configure point clustering, look at different hatch and transparencies so that we could bring things like your design data or your infrastructure data back in over the highest level resolution that Nearmap offers through Map Browser.

And then, the last thing I want to talk about is actually doing the inverse, so leveraging the connectors for ArcGIS inside of Autodesk product. So Autodesk continues to improve the connectors for ArcGIS in Civil 3D, in InfraWorks, in Map 3D. And these can be a valuable way to leverage Nearmap data inside the Autodesk desktop products, as well.

So we talked about before, publishing out your design data and bringing it into Map Browser. This is the inverse of that, of taking the Nearmap data, publishing it out to ArcGIS Online or ArcGIS Enterprise, and bringing it into your Civil 3D or InfraWorks desktop application.

So to be able to connect to Nearmap data through ArcGIS Online or Enterprise, to provide access to your private or subscription services for your project using ArcGIS to host the data, to access public data as context to imagery and your design environment, to add feature layers to your Civil 3D as objects like COGO points, alignments, pipes, and networks.

And we can add that through points, lines, polygons, and raster data to be able to publish that data to ArcGIS Online or Enterprise to leverage the design data back into Nearmap browser, like we showed in the last slide and additionally, to be able to update and save changes to feature services, attributes, and geometries back into the services directly from your Autodesk desktop or even add new records to existing feature services.

So there's a lot going on with what we can do with the imagery and the imagery products. So at this time, I'm going to hand it over to Jeff to talk about applying imagery products.

JEFF SAUNDERS: Thanks, Steve. That was great. So you heard Steve talk about the project life cycle and the key stakeholders involved in an infrastructure project. And then, you also heard about all of the imagery products that we bring to bear on this market to support the infrastructure space.

What want to talk a little bit more about now is how that combination of the imagery data products and the tools from Autodesk really help to benefit AEC professionals as they look to develop and initiate projects looking at additional or existing conditions in context and then using that information as they plan, design, construct, and manage throughout that project life cycle. So we're going to walk you through some of the ways that imagery, on the next slide, begins to be leveraged in each of these different tools.

So Steve highlighted a lot of these already. But the 3D Textured Mesh, the Point Cloud, the Digital Surface Model, the Digital Elevation Model, True Ortho Imagery, and the resulting AI-based features that are extracted and planimetric data that can be derived from that-- all of those data sets are really plug-and-play content. As you begin to start an infrastructure project, you can leverage this information at any point around that project life cycle, as Steve highlighted in one of the first slides.

And that data goes straight into tools like InfraWorks, or Recap, Civil 3D, 3ds Max, Revit, Twinmotion now. And there's plenty of others. And we're going to spend a little bit of time showing you some of the examples that we continue to get excited about as our customers develop-- our joint customers, I should say-- develop and deliver on their infrastructure project workflows.

So the next slide, this just speaks to how a point cloud from our imagery data products can be used to look and assess and establish the existing conditions for a new project. This is just panning around and viewing it. But because of the level of detail and the large scale that this imagery and content is captured at, users-- both stakeholders across the life cycle-- can benefit from these views as they're talking and sharing information with other stakeholders in the project.

The next one actually is an interesting one because we also see a lot of our content being used in different visualization engines. So by leveraging 3ds Max, we're able to bring in meshes and point clouds in a way that can then be leveraged further on in the development and conceptualization life cycle to provide context to provide additional content for those projects as they move, as they're shared and communicated with other stakeholders.

This is just one of the examples of how some of our joint customers are starting to build immersive experiences to really engage the stakeholders early. Steve talked about a picture being worth a thousand words. The more that we see these models being used to communicate project plans and infrastructure projects at the earlier stages and proposals and planning, the better that we see the understanding of what those projects will look like, how they'll affect the communities, and how they improve on different aspects that an infrastructure project can support.

But it's not only the immersive experiences. It's also the beginnings of more and more use cases for simulations. And a lot of these simulations need to start with reality data, data that can provide better context and better content to frame an infrastructure project as it's being described and communicated and explained to key stakeholders.

So it's not just a nice flyover. It's now becoming more and more of a simulation, a way to simulate how that new experience with new infrastructure will operate within a city, within a neighborhood. And this is just some more exciting pieces that we see our customers starting to use our content with the Autodesk tools to develop.

Now, sometimes, the simplest pieces are the most important. And almost every preliminary design can always benefit from the most current topography data that's possible. Steve talked about the DSM and the DTM that we generate. And that content, similar to the others that we've been talking about, is plug and play.

Drop it into a Civil 3D environment, generate the contours, and you have highly current topography to really start and develop your preliminary design against. So that will give you rich context and appropriate terrain as you're looking at preliminary designs, even conceptual designs in some cases, to provide that foundational content.

But the data doesn't just go into Civil 3D, InfraWorks and similar tools. It's also very valuable in the context of a Revit model. And the ability to, again, provide that context for new designs and how to cite projects within those, that context is really critical.

We continue to see really exciting ways that different AEC professionals are leveraging the 3D mesh, the 3D point cloud in Revit models as a way to provide a starting point and also as a way to understand what the environment around a new project will look like.

So some of the other things that we've been excited to see over the course of this year has been further adoption of Autodesk Construction Cloud and our data being pushed into Autodesk Construction Cloud-- and then, in conjunction with ArcGIS GeoBIM, creating dashboards, creating project tracking, project management related efforts, bringing all this data together in a way that this information, both on the geospatial side and on the construction and engineering side, come together in a consistent view that allows different stakeholders to interact with the information in unique ways and really leverage this reality data in coordinated workflows across all the disciplines. And this is just one example of how we've seen that happen.

As more and more infrastructure projects happen, our library of historical data continues to help. And as we continue to fly new projects, we continue to develop this catalog that allows customers to look at both historical as well as this idea of tracking project progress over the span of some of these larger, longer time frame projects.

So as you see on the right-hand side, there's a project where we've captured imagery across the whole project life cycle and watched it move from the grading to the actual construction and final handoff in this whole time frame.

Sometimes, even more exciting is watching, is being able to take a look back and understand where a current city or neighborhood came from by looking at the evolution and changes in the contours and the city layout over time. So this is just, again, a time track over historical imagery that allows provides better site condition awareness as well as being able to show off the development of a city, of a neighborhood in ways that provide some unique views for key stakeholders in the process.

STEVEN SANTOVASI: So I know that we've just given you a whole bunch of information. And we've overloaded you with demonstrations and looks at products. We want to look at this as the big picture. What's the takeaway here? And I think that the big thing is that imagery products add context to a project.

And the better data that you have, the more insights you're going to have. So if we look at this, the different types of imagery products that we create from these aerial flights, including all the 2D products, all the 3D products, all the AI products-- essentially, they give you deeper insights. They help you identify, throughout historic data, what a project site looks like.

They help you monitor the construction across time. And they give you the ability to keep the project safe and environmentally clean and successful in the end. And what we hope that you take away is that Nearmap data can help throughout the entire life cycle of the project.

So bringing Nearmap and AEC solutions together, what does that mean for you? It means plug-and-play geospatial and reality data. It means, like I said, shortly after we fly it, we turn it around and we deliver that data that you can then immediately bring into your project to better engage stakeholders and proposals and design stages in addition to construction turnover and operations.

Context for existing site conditions will help you jump start your project. And you can have savings in the order of hundreds of thousands of dollars per project. In many cases, we've talked to firms that are saving as much as two FTEs by using Nearmap data, by reducing time consuming digitization to reduce field visits and with the automatic refresh of data that takes place two or three times a year.

This helps them increase project revenue on their EPC projects and the use of current and historic data helps them to identify change over time and to monitor the built environment. So I think with that, what we're going to do is we're going to start to answer some questions, have a little bit of an interactive discussion and some observations on what we presented today.