Automated cartography using AI

Your feed might be dominated by LLMs these days, but there are some amazing things happening in computer vision that you shouldn’t ignore! In this episode, we bring you one of those amazing stories from Gabriel Ortiz, who is working with the government of Cantabria in Spain to automate cartography and apply AI to geospatial analysis. We hear about how AI tooling fits into the GIS workflow, and Gabriel shares some of his recent work (including work that can identify individual people, invasive plant species, building and more from aerial survey data). Sponsors: Fastly https://fastly.com/?utm_source=changelog – Our bandwidth partner. Fastly powers fast, secure, and scalable digital experiences. Move beyond your content delivery network to their powerful edge cloud platform. Learn more at fastly.com https://www.fastly.com/?utm_source=changelog&utm_medium=podcast&utm_campaign=changelog-sponsorshipFly.io https://fly.io/changelog – The home of Changelog.com — Deploy your apps and databases close to your users. In minutes you can run your Ruby, Go, Node, Deno, Python, or Elixir app (and databases!) all over the world. No ops required. Learn more at fly.io/changelog https://fly.io/changelog and check out the speedrun in their docs https://fly.io/docs/speedrun/. Typesense https://cloud.typesense.org/?utm_source=changelog – Lightning fast, globally distributed Search-as-a-Service that runs in memory. You literally can’t get any faster! Changelog News https://changelog.com/news – A podcast+newsletter combo that’s brief, entertaining & always on-point. Subscribe today https://changelog.com/news. Featuring: Gabriel Ortiz – LinkedIn https://www.linkedin.com/in/gabriel-ortiz-gis-cantabriaDaniel Whitenack – Website https://www.datadan.io/, GitHub https://github.com/dwhitena, X https://x.com/dwhitenaShow Notes: Automated cartography (integration of different models: buildings, roads, vegetation) https://geocontenidos.maps.arcgis.com/apps/webappviewer/index.html?id=6b923b724cba46efb37411fdd89ec9faDetecting and tracking the expansion of forests (period 1957-2020) using both legacy and modern imagery https://storymaps.arcgis.com/stories/1db87a67f3614cc4b58f7e00f44af03eTracking invader species https://cantabria.maps.arcgis.com/apps/webappviewer/index.html?id=a7305e3e80394d769ff789bc6c4909c4Tracking urban growth with AI https://experience.arcgis.com/experience/ba08444d130a47d4835b6cf2ddd2049aSpatial behavior in beaches using AI https://storymaps.arcgis.com/stories/862bf11ce2034208b47ef43f32a7e84aInference with SAM (Meta’s Segment Anything Model) over urban areas https://geocontenidos.maps.arcgis.com/apps/mapviewer/index.html?webmap=921f134ac43240df8317d8d26d0caff1/viewer.html?webmap=4af373c294e24394ae25e4acadab71ccSuperResolution on aerial or satellite imagery https://cantabria.maps.arcgis.com/apps/webappviewer/index.html?id=f14ab31439644f118c3ca7ef8c2258c9More of the work of Gabriel and his team can be seen here https://mapas.cantabria.es and also on his LinkedIn profile https://www.linkedin.com/in/gabriel-ortiz-gis-cantabriaUpcoming Events: Register for upcoming webinars here https://practicalai.fm/webinars!

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:07Speaker 00

  Welcome to Practical AI. If you work in artificial intelligence, aspire to, or are curious how AI-related technologies are changing the world, this is the show for you. Thank you to our partners at Fastly for shipping all of our pods super fast to wherever you listen. Check them out at fastly.com. And to our friends at Fly, deploy your app servers and database close to your users. No ops required. Learn more at fly.io.

• 0:42Daniel Whitenack?

  Welcome to another episode of Practical AI. This is Daniel Whitenack. I'm a data scientist and founder at Prediction Guard. And I am really excited today because my, my life has been filled with large language models for, for the past months. And I feel inundated with information about those. But there's so much going on, so many amazing things happening in the AI world outside of the text modality. And today we have with us Gabrielle Ortiz, who is Principal Geospatial Information Officer at the Government of Cantabria in Spain. Welcome, Gabrielle. How's it going?

• 1:24Speaker 02

  Thank you. Thank you. Thanks for having me on and giving me the opportunity to share with you and the audience what we have been up to in the last few years regarding your spatial analysis and particularly artificial intelligence.

• 1:37Daniel Whitenack?

  Yeah. And one of the things that stood out when, when we started talking, well, first of all, you're, you're a listener of the show. So I love it that you now get to be a guest on the show. That's so wonderful. I'm glad that we have listeners who are doing amazing things as practitioners. But also you're, you're in Spain, which is, is one of my favorite places. I did my collaborators during my grad school days were in San Sebastian and I spent time there. And I know there's so much innovation happening in that region and in Spain in particular. What's it like to be working in AI in Spain?

• 2:14Speaker 02

  Well, Spain is, I think is a great country to find professionals in all the branches of engineering and There are many things happening in the AI industry. There is a lot of good, you know, environment of startups growing. And I really encourage you to engage and contract people from Spain.

• 2:38Daniel Whitenack?

  Yeah, that's so great. And not only is there amazing work going on there, but it's one of the most beautiful places I've I've been. And even when you logged in, so our listeners can't see it, but I see the beautiful sunshine and trees and and town behind you through your windows. So I'm I'm a little bit jealous.

• 2:57Speaker 02

  You mentioned San Sebastian. I am pretty close to San Sebastian. San Antonio is a really, really beautiful city.

• 3:05Daniel Whitenack?

  Yeah, yeah. So we mentioned that you work in geospatial. I know, so I've been on the mapscaping podcast a few times, which has been has been fun. And I know that that industry is really wrestling with kind of uses of deep learning, uses of AI and understanding how to integrate that into workflows. If my understanding is right, you didn't come from a data science researcher sort of background into this topic. You came more from the geospatial side. So could you tell us a little bit about how, as a geospatial practitioner,

• 3:47Daniel Whitenack?

  you first started kind of dipping your toes into deep learning and understanding what it meant for for your industry?

• 3:56Speaker 02

  Sure. I have been working in the special industry for more than 30 years. I started working for topographic control, pathimetric control of works. Then I move on to engineering companies designing highways and railroads, dealing with environmental data and always using GIS, which stands for geographical information system. And as many of you know, it's a technology that lets you operate and do analysis over a huge amount of data. Then I started to work for the government of the Cantabria. I am now in the role of Principal Geospatial Officer, as you mentioned.

• 4:39Speaker 02

  But literally, if you translate directly from Spanish, it would be something like Chief of the Service of Cartography and GIS. My role here is not only being in charge of the data production, but also in the development of the infrastructure for the analysis of your spatial data within our organization. It means for our staff, but also for our stakeholders outside, which is something very important for us, for the citizenship, for the community and for the companies that are working with your spatial data. Me and my team, we have something very ingrained in our DNA, which is the public service that we provide using AI

• 5:24Speaker 02

  and using another set of technologies. And we every day try to do our best to fulfill with that target.

• 5:31Daniel Whitenack?

  Yeah, it's really, really inspiring to hear kind of the motivations behind how you think about doing your work and the people you're serving, which is so great. I'm wondering, just practically, you mentioned kind of GIS tooling and, you know, the processing of data in that space. Of course, deep learning and the AI space has its own sort of unique tooling and sometimes weird tooling. So I'm wondering, could you comment in terms of, what is it like for a geospatial practitioner to start adopting deep learning techniques and all of that,

• 6:14Daniel Whitenack?

  which I'm assuming have a different set of tools than geospatial people have used in the past? So what is the current state of, you know, the tooling around mixing deep learning with geospatial? Is it difficult? Is it fairly segmented? Or is it more integrated at this point?

• 6:33Speaker 02

  Well, at first sight, it seems daunting and intimidating. But I have to say that it is not so difficult, right? Just to demystify a little bit the AI technology. As you mentioned before, I am not a researcher on AI. I am an expert in the spatial industry. And I will tell you my story, how I began. My first, you know, contact, or at least the first time that I pay attention to AI, was in 2012 with AlexNet, what happened in the ImageNet challenge. At that point in time, classification of images was great, but it was not very applicable to the spatial industry. You know, it has an application and you can leverage that,

• 7:17Speaker 02

  but it's not what we do every day. Previous to that, I have to say that it was in 2010 or 2011, something like that. I knew about the work of Nvidia with the GPGUs, the general purpose GPGUs. I think Bill Dali talked about this in one of your early episodes. And that was very interesting for me, because in the spatial industry, we often have a lot of demanding in terms of computing power. When we operate with what we call raster data, which is no more than data organized in a grid, topologically in a grid. For instance, an image is raster data. But also, for example, a digital terrain model,

• 8:01Speaker 02

  which is a grid where you store it in every pixel in the center of every cell, the value of the altitude of the terrain over the main sea level, for example. And you perform calculations over that digital model. For instance, you know, getting the watershed or the viewshed of one part of the territory. And that calculations can span for several days or even weeks. Because in spite that the mathematics underlying, running under the hood are not very complex, what happens is that you have so many pixels that it ends up being very demanding.

• 8:40Speaker 02

  And what NVIDIA started to do on those days was to be able to parallelize a lot of calculations and instead of using four computational threads on your GPU or eight computational threads, they were able to spread all the calculations among hundreds or even thousands of computational threads. That caught my eye because it was very important for me. But at that point in time, I thought, you are going to need, Gabriel, you are going to need a GPU, but not for artificial intelligence. I was not thinking about that, but for calculations of a different nature.

• 9:17Speaker 02

  Then in 2015, 2016, we witnessed the blossom of a whole new generation of deep model architectures. Just to mention some of those who had a big impact in computer vision. ResNet in 2015, I think it was presented in 2016, I'm not very sure. Then U-Net, that has been extensively used. In 2017, the Facebook Artificial Intelligence Research Group presented and proposed the Mask R-CNN, you know, it was an evolution of Fast R-CNN. And in 2018, I saw for the first time a demo within the spatial realm of our data provider,

• 10:02Speaker 02

  which is Esri, actually. I think you also have a couple of guys from Esri in a previous episode. And what they were demonstrating is how you can detect swimming pools and oil rigs automatically using a single shot detector in that days. And that was kind of an aha moment for me because I realized, well, you have to invest your time, this is going to be definitely a game changer, and you have to start working on this. So that was the moment and from that point, you know, there are two kinds of persons. I will use a metaphor to explain that when you see the results of AI,

• 10:45Speaker 02

  some people think it's magic and everybody likes magicians, you know. Some people end up falling in love with the magician. You know, they are obsessed with the persona and the mystery and the whole stick. But some other people just want how the trick is done. And I think I belong to the second group. So it was not only that this looks like magic. The point was how this is done. And from that point, I started to work. We can delve into this if you want, but this is not so difficult as I said before.

• 11:24Daniel Whitenack?

  Yeah, that's so great. Yeah, I think I applaud you for digging in and, you know, not too early where it was only a research topic, but as it started getting into practical applications, you really took that and figured out how to apply it within your context appropriately, which I think is maybe not everybody takes that approach. So I appreciate that. So with the tooling that you're using, I think maybe this is useful for people that haven't done geospatial as much. So I know there's major tools like ArcGIS and other ones. And then you've got sort of like Jupiter notebooks where you,

• 12:09Daniel Whitenack?

  you know, train models or GPU services where you can run inference and other things. Have those merged at all? So like from within the tooling that you're using as a GIS professional, has some of the deep learning tooling been integrated into those tools? Or is it mostly at this point for you, I'm going to export my data from the geospatial side and then use a notebook and then import it back in or something like that?

• 12:38Speaker 02

  Well, that's a smart question. As I said before, our software provider is Esri. We worked with Esri for a number of years and they are doing an excellent job in integrating many open source frameworks into their platform.

• 12:56Speaker 02

  And I think, because we try to follow the literature, but we are constantly falling behind, you know. It's extremely difficult every week.

• 13:04Daniel Whitenack?

  Me too.

• 13:05Speaker 02

  Every month. It's impossible. And even, you know, constructing and completing the puzzle and the issue of installing all the frameworks and then putting everything into work can be very complex. So we have a big advantage working with the S3 technology. They have a team, research and development team based in India. And I think these people is doing a great job facilitating the application of that. In some of your previous episodes, you have been talking about UX interfaces for using artificial intelligence, whether or not it has, it makes a difference.

• 13:48Speaker 02

  And it really makes, because it's a way of, you know, democratizing and making accessible the technology. That is one part of the story. I think it has facilitated a lot our work, because you not only need the frameworks, you need all the platform to move across terabytes of data. Your special industry is highly demanding in terms of the data that you have to work. And it's not only the frameworks of open source. It's how you prepare the labeling, how you structure the databases. There is a lot of more science. And also, apart from that, what I did is starting to gain the main concepts

• 14:32Speaker 02

  related with artificial intelligence from all the great resources that are completely free on the internet. You know, on YouTube, you have lessons from the MIT, from Stanford, that can introduce you to the simplest concepts such as a perceptron or a back preparation or stochastic gradient descent. So I designed for myself a two-fold strategy. First, trying to gain experience on, with getting hands-on of the shelf models. But at the same time, trying to also learn about the concepts underlying.

• 15:10Speaker 02

  Pinpointing the AI world. I think that's important. Many people think that artificial intelligence is a black box. It's not a black box, it's mathematics in action. Of course, it's not linear. You cannot fully predict what is going on, but many of the things can be understood.

• 15:30Daniel Whitenack?

  Well, I love your perspective, Gabrielle, on how you've developed a mental model of how these technologies work. I think that's an encouragement to others to both explore these technologies, but also keep in mind what they are and how they should interact with them as tools. But I'm so fascinated by some of the projects that you've been able to accomplish during your time using this technology. And I want to start diving into those a little bit. One of the ones that you pointed me to that was really fascinating reminded me of standing on the beach in San Sebastian,

• 16:12Daniel Whitenack?

  although it looks like you have really maybe more nice beaches up where you're at. So tell us a little bit about how standing on beaches and counting people on beaches, why is that important? And how did you get into this project of applying deep learning in that context?

• 16:31Speaker 02

  Yeah, definitely. I started working with deep learning at the end. I think it was the end of 2019 or something like that. Then came the pandemic. And after the pandemic, you know, with the release of restrictions, somebody here at Government of Cantabria said, hey, we are a little bit worried about the possibility of having uncontrolled crowds on the beaches. Because I have to say that Cantabria is a notable tourist destination. We have more than 100 beaches. So you can have a big problem in terms of spreading of the COVID-19. And they were worried. The first thing that they asked to me is,

• 17:11Speaker 02

  how can we get a calculation of how many people we have in every beach? And when the tide is up and when the tide is down, things like that. But it was just a simple calculation in terms of the surface or the area that the beach has. And I said, "I think I can go further, and I will count the people." And they said, "What? Are you crazy?" "Yeah, I'm not drunk, I think I can do it." Because I had some experience using single-saw detectors, and at that point in time, more models than single-saw detectors. And it's what we did.

• 17:53Speaker 02

  We started counting the people, because normally we have an archive of aerial, you know, surveys conducted, always, as is normal, in clear skies, sunny days, when everybody is on the beach, in summer. So we had very well the behavior of use of every, especially the behavior of use of every beach all across Kandabria. At different days, different months, different, no matter if it is a weekend or a Labor Day, we had a huge amount of data to analyze. And we developed some deep learning models, that even if you are changing the input signal,

• 18:35Speaker 02

  that means changing the aerial survey, it works. We could predict the sectors of every beach, not only in terms of absolute figures of population in a beach, but which are the sectors where the people try to concentrate. And after that, we released a small application, that you can see in the notes of the podcast, where you can see some maps. And if you, just for curious interest, if you want to a place, "I want to go to a beach, and I would stay quiet and loosey-goosey, you know, without many disturbances," you

• 19:18Speaker 02

  can see what places are the most suitable for that use. So it was a great experience, our first experience releasing something.

• 19:27Daniel Whitenack?

  Yeah, that's so fascinating and it makes so much sense after you say it. I know here I can think of so many more applications for something like this. I know like in the in the US national parks, you know, thinking about crowding and the impact on the natural environment or other things like that and helping plan out for crowds at certain points of the year. There's there's so much practical use of this and this was amazing because, yeah, you took this knowledge that you had been building up and really applied it in the moment during COVID-19 when there was this specific need. But then it sounds like there's continued usage past

• 20:08Daniel Whitenack?

  that because even if I'm just a consumer, like I'm a normal citizen and I want to enjoy the beaches, this information is really useful to me. I I know myself, I probably would go to the quiet places of the beach and

• 20:21Speaker 02

  Yeah.

• 20:22Daniel Whitenack?

  and sit and listen to the to the waves. So that's

• 20:26Speaker 02

  There are much more interesting problems to try to solve than the one that I described.

• 20:33Daniel Whitenack?

  Yeah.

• 20:33Speaker 02

  And now later, we started to work trying to modelize or to model certain aspects of how the territory works. You have to understand the territory as a whole, as a living entity, which where everything is related to everything. So we started to slice every variable and try to address those variables with the help of AI. For example, we have developed some interesting models. We can delve into the, you know, the architectures you want used later on or whatever you have interested in. But some interesting models for detecting and classifying vegetation.

• 21:17Speaker 02

  Also for the evolution of urban growth, also for things like where like tracking cars, for example, that is like a kind of proxy of the society, how the society moves. And because everything is on our aerial surveys, you only have to have the skills to bring back that information and convert it in something useful. And we have been, as the years went by, we have been able to produce some more relevant results. I will not talk about deep learning models, but about solutions for tracking the territory.

• 21:55Daniel Whitenack?

  Yeah, and you've mentioned aerial surveys a couple of times. It may be useful for those in our audience who don't work in geospatial. They might have in their mind maps and things like Google Maps, where, oh, I could go and I could look at a satellite image, but it's not current, right? It's maybe one photo that was taken some while back. And you've talked about aerial surveys, where you can actually learn, you know, both current information about what's going on in an area, but also historical information. Could you just help our audience understand, like, as a professional, what sort of data do you have access to? And how is that gathered practically and made available to you?

• 22:37Speaker 02

  Well, I have to say that everything that I have been talking about can be also executed with satellite images, you know. There are some differences, but of course, you can do it with satellite images. The reason that we work more with aerial surveys is because we are more focused on capturing this kind of information rather than working with satellite data. My region, Cantabria, is not very big. And we have in Spain a national plan that covers every three years all the country with aerial surveys. And also, we have a repository of satellite images. So anyway, you can do both of the input signals.

• 23:22Speaker 02

  The results will differ slightly. But apart from image capture with sensor, no matter if it is airborne or satellite sensor, we also work with a range of technologies. For example, LiDAR data. I know that many of the audience have been working with LiDAR data. LIDAR can be also airborne. In fact, it was the origin of the technology, LIDAR, using from a plane, you know. And it has been increasingly important in our domain. We also work with, you know, a system of records with traditional databases and a number of things.

• 24:06Speaker 02

  If I had to say something about my job is that is extremely interesting because one day we are working with COVID data, for example, another day you are working with energy data, another work with environmental data. The government of Cantabria has powers and duties in many domains. It's kind of one of your estates. If you forget the difference of area cover, only Texas or Florida. I think Spain is in the middle between the area of Texas and Florida is something in between, but the whole country. And my region is quite small, but it's a very interesting place to work with because that reason.

• 24:51Speaker 02

  And the data comes from many different technologies and many different databases.

• 25:03Speaker 00

  This is a change log news break. OpenObserve is a cloud-native observability platform built specifically for logs, metrics, traces, and analytics designed to work at petabyte scale. Huge! According to its creators, quote, "It's very simple and easy to operate as opposed to Elasticsearch, which requires a couple dozen knobs to understand and tune. With OpenObserve, you can get up and running in under two minutes. It's a drop-in replacement for Elasticsearch if you're just ingesting data using APIs and searching using Kibana. Kibana is not supported nor required with OpenObserve. OpenObserve provides its own UI, which does not require separate installation, unlike Kibana."

• 25:48Speaker 00

  End quote. An interesting offering indeed. Here's a couple choice quotes from the comments section.

• 25:56Speaker 00

  User GetTutachapa says, "I just tried this three days ago. As someone running the home lab and hadn't set up logging yet, it was a great find. I didn't have to learn and combine three plus log technologies. It's just a single all-in-one monitoring server with web UI, dashboards, log filtering slash search, etc. RAM usage of the Docker container was under 100 megabytes." And user Surge Axe says, "Interesting product. Thank you for the effort. Definitely want to give it a try. For me, though, setting up a system is not the primary pain point today. For what it's worth, signing up for a cloud service is not hard. The problem starts at the ingestion point."

• 26:40Speaker 00

  You just heard one of our five top stories from Monday's Change Log News. Subscribe to the podcast to get all of the week's top stories and pop your email address in at changelog.com/news to also receive our free companion email with even more developer news worth your attention. Once again, that's changelog.com/news.

• 27:07Daniel Whitenack?

  So Gabrielle, we talked a bit about this kind of first project of related to population and crowding on beaches, but you've done so much more. Could you highlight a few of these things in terms of other, other things you've been able to identify or track with deep learning from these aerial surveys?

• 27:28Speaker 02

  Yeah, we have an extensive work in the detection of vegetation. I have to say that we have been also only using supervised learning, that branch of the deep learning, and specifically working with different model architectures such as I mentioned before, UNet, you know, Mascar, CNN.

• 27:53Speaker 02

  And some others. Now we are testing now some segment anything model, but we haven't done anything with zero shot learning for production. So what I am going to tell has been achieved using model architectures that have been almost forgotten for the community. You know, everybody is focused on the Sora architectures and there are so much that can be extracted from the old school of artificial intelligence.

• 28:22Daniel Whitenack?

  Yeah.

• 28:23Speaker 02

  Quote unquote, it's not so what, right?

• 28:25Daniel Whitenack?

  Yeah, yeah. And I think actually, this is maybe a misconception of people that occasionally we try to mention this on the show, the majority of applications across enterprise, not just in GIS, but in manufacturing or marketing even, people think of marketing with generative AI, but the majority of applications are still traditional, quote, traditional machine learning. You know, there's a lot of scikit-learn models out there still, or there's just supervised learning, you know, models out there. And yeah, it's awesome to see, to highlight that here because I think it is a misconception.

• 29:03Speaker 02

  Yeah, because you know, when a paper appears, normally they do not run out the possibilities of the model. The professionals who are not very specialist in the AI domain, but are have a lot of knowledge in a specific domain out of AI, in my case, we can prepare and curate better labels. We can understand the process that we are trying to model. And we can, we have so much to give and to propose to the, to the community. And that's one of the reasons that some people have said that your models are quite good. How have you

• 29:44Speaker 02

  done it? Is a brand new architecture is something that you have created on your own? And always say, no, it is not. It's just using in a smart way, model architectures proposed back in 2015, 2016, but with a lot of data, very well created. I also have to say that the computer power that we have at our disposal is quite modest. We don't have from that point, something, you know, big or.

• 30:13Speaker 02

  Are very extensive and the key is how you create the data.

• 30:18Daniel Whitenack?

  Yeah, and one of the things that you had mentioned prior to recording was this idea of automated cartography as kind of an integration of a bunch of these different models that that you've been working on. I'm wondering if you could kind of first describe what do you mean by automated cartography? And maybe even for people that aren't familiar, what is cartography? And, you know, I'm assuming modern cartography isn't like, you know, Magellan getting on his paper and drawing on drawing, you know, maps on parchment paper or something. But what what is cartography look like these days?

• 31:00Daniel Whitenack?

  And then what do you mean by sort of automated cartography with these sorts of models?

• 31:07Speaker 02

  Well, cartography is the art and the science of, you know, trying to model the reality and abstract the reality and plot it in a flat surface. It's a science that has been developing for a number of centuries, from, you know, many centuries. And up until now, it was highly dependent on the human, you know, ability to trace and and to, you know, to draw everything on the surface of the earth. As the technology has been developed from the 90s, we started to move very rapidly into digital technologies. And the automation of the cartography has taken place not only with the advent of AI,

• 31:52Speaker 02

  but several decades before. However, this is a revolution because we have never been able to is to produce high such a degree of quality using so few people working. There are some similar technologies like remote sensing, which is the the part of the technology in charge of analyzing from imagery of satellites and, you know, producing cartography also. You know, it recalls many things of the artificial intelligence, but it can match the results in many other fields. So the revolution continues.

• 32:31Speaker 02

  It started, as I said before, in the 80s, 90s, but now is a complete revolution. And I think that For the first time, we are able, we have an example that you can check it out in the description of the podcast, but where we have been able to produce a map with a basic coverage, where you have trees, where you have shrub, where you have no vegetation, where you have buildings or roads or railroads, completely generated by AI. Of course, it has some mistakes, but we left on purpose, those mistakes on purpose, because

• 33:11Speaker 02

  we wanted that the rest of the community could evaluate the capacity and the ability of the models to work alone.

• 33:19Daniel Whitenack?

  This is a question that just popped into my mind as you were talking about these models, what's possible, and you know, it's not perfect, right? No AI system is perfect, so there's going to be mistakes. I'm wondering, as someone who's been in GIS and been a practitioner for, I think you said 30 years now, I also imagine that human-based processes are error-prone, or at least they're slow, right? So by the time a human maybe processes a certain map or something, things have been updated, right? And it's maybe not current anymore. What do you think about the, what are the implications for maybe cartography or GIS

• 34:04Daniel Whitenack?

  as we move to the future where AI systems maybe can do things more up-to-date, but with some mistakes, but they're up-to-date and can really maybe highlight certain areas that are incomplete or something? Combined with human efforts to correct those mistakes and keep the, what do you see as this balance between trying to be automated with AI-based techniques and the role with that human cartographers or GIS professionals play as these systems expand to more and more places?

• 34:42Speaker 02

  Yes, it's a very interesting question, because you know, one of the big problems that we have is to maintain up-to-date every single database that we release into the market for our stakeholders. That's a very big problem, because it's always difficult.

• 35:01Speaker 02

  And one of the main advantages of artificial intelligence is that you can have a model and you know. It will not probably work perfectly with the next area survey because it will have some differences in terms of colors or shadows or whatever, but you can fine tune. You can, or maybe you can train the model from scratch again. Start from scratch with the training and you can update something in a reasonable time frame. So that is one of the things that I most attracted by, the capacity of updating things. And it's a game changer, as I said before.

• 35:43Speaker 02

  No, no other artificial intelligence offers things that other technologies really don't.

• 35:50Daniel Whitenack?

  Yeah, and of course there's limitations. You know, AI is never, the expectation should never be that it solves all of our issues. But it also should be that it, you know, it's going to solve some of our issues or solve some of our problems, but not all of them. From your perspective, how do you think about the current limitations of AI within GIS and cartography? What are some of the things on your mind with respect to that?

• 36:19Speaker 02

  Oh, yeah, I think of course you have to bear in mind that we have limitations. What happens to me also happens in teams in India or in the US that I am always seeing what they are doing. I would like to point out two limitations. One is the computing power and another thing is the limitations of CNS, which is the technology that we are using right now, convolutional neural networks, right? We can talk a little bit about model architectures and things. In terms of computing power, I think it's worth delving into the role of GPUs because, because in the spatial realm, it's not well understood, why do we need a GPU? And

• 37:05Speaker 02

  It's something, I don't know if it happens in other markets, but in our industry, when you talk to somebody about a GPU, normally my fellows and mates, I don't know, they try to say it's something related with the IT department. I don't want to be in charge of that, but it's not at all. You have to be aware of what technologies do you have for calculation. The hardware is so important and you have to speak the same language as a data scientist that the rest of the community speaks. And that is very, very important to understand that it's not the same a GPU in your laptop that DGX or talking about the, or H100,

• 37:50Speaker 02

  if we are talking about Nvidia hardware, it's not the same. And it's everything related with the amount of data that you want to put into the train. You know, the quality of your training, the level of the convergence that you are going to get, if you are going to stay in a local minimum in the convergence or you are going to reach and source the possibility of the label that you are ingesting into the model. Everything is related with the hardware. I think that Bill Dali and Animan and Akuma in many of their talks always talk about the trinity of AI. One of them is, you know, the data. Another is the software, the algorithms, that many of them have been with us for a while.

• 38:34Speaker 02

  Propagation, you know, many of those algorithms have been from the eighties, if not before. But the hardware is the third part. Bill Dali always says that it's the spike that starts this engine of creativity of AI. And I think it's true. You have to pay a lot of attention to computer power. And there is another limitation that is ingrained in the DNA of the CNNs. As far as I know, from my experience. You cannot expect to perform exactly as a human being. And sometimes, in spite that you create very well your labels and your tips and your data, the model do not learn as good as you expect.

• 39:19Speaker 02

  But somewhere in between, you can have a reasonable amount of success in that. What we do to overcome is combining different model architectures. It's something very useful and widespread in the spatial industry. For instance, we combine models at two different levels. From the architectural levels, it's quite common to see the combination of ResNet with, for example, U-Net. In ResNet, you remove the last part, the fully connected layers, and connect the remaining part with U-Net. So you are using ResNet as a feature extraction, for feature extraction,

• 40:03Speaker 02

  and then the rest of the decoder happens during the rest of the U-Net architecture. It also happens with the MARS RCNN. We use constantly ResNet as a backbone, but then the rest of the model goes. And there is a second point, which is combining the results of, or the inference, when you have inference from two different model architectures. For example, talking about vegetation. Imagine that you have one model that detects very well the big areas of vegetation, but fails in the small spots. And you have another model that works very well for small spots, but fails detecting the big areas,

• 40:46Speaker 02

  because in the big areas creates artificial holes and mistakes. You can combine the results of, the outcomes of those model architectures with traditional GIS techniques to mesh all the results together and obtain a bigger, best quality of the layer that you want to infer. That has worked for me and is one of the ways that we are trying to overcome the... The limitations of artificial intelligence.

• 41:18Daniel Whitenack?

  That's great, super practical. And I know that's what a lot of our listeners want to hear is some of the practical ways they can explore these technologies. Well, Gabrielle, it's been an amazing pleasure to talk to you. As we close out here, there's a million things we could talk about. I know some we didn't get to, and we'll link in the show notes. But as you look to the future, could you just briefly, in the last minute or so, just briefly share with us, what's exciting for you as a GIS professional looking to the future that either you want to dig into next, or what are you encouraged by or optimistic about as you look to the future of your own work and how AI influences that?

• 42:02Speaker 02

  Well, I have to say that in my 30 years plus of working in the spatial industry, these two last years, two or so, have been the most exciting part of my career, because it's so creative. We are just scratching the surface of AI, and great things are coming. I think that with the advent of Shiro Shot, we have been watching from the first week of April, what can be done with the SAM, the Segment Anything model. And I'm sure that new versions will come of future versions of SAM. When we combine that with LLMs, with large language models, and we can interact

• 42:46Speaker 02

  with the voice and say, hey, draw me all the trees in the image, or it will be much easier to use this set of technologies. Anyway, just to finish, I would like to send a message to the audience, for those who are not artificial intelligence researchers like me, that it's possible to apply this set of technologies, even though you are not an specialist on that specific domain, is also to get hands-on on one of the, take one of the shelf models and start playing around with them. And I know the future will be absolutely focused on artificial intelligence.

• 43:29Speaker 02

  There will be a different geography in the next few decades.

• 43:35Daniel Whitenack?

  Awesome. Yeah, this is so inspiring. Thank you for your work, Gabriel, and it was awesome to have you on the show. Thank you so much.

• 43:41Speaker 02

  Thank you.

• 43:51Speaker 00

  Thank you for listening to Practical AI. Your next step is to subscribe now if you haven't already. And if you're a longtime listener of the show, help us reach more people by sharing Practical AI with your friends and colleagues. Thanks once again to Fastly and Fly for partnering with us to bring you all Chang Talk podcasts. Check out what they're up to at fastly.com and fly.io. And to our beat freak in residence, Breakmaster Cylinder, for continuously cranking out the best beats in the biz. That's all for now. We'll talk to you again next time.