The Alamo! 3D Laser Scanning and Photogrammetry to Create a 3-Kilometer Computer Model

LARRY KLEINKEMPER: What countries do we have represented here? So the people who put their hands up for Europe, can you just give us some ideas? You can just yell it out. You don't have to put your hand up.

AUDIENCE: [INAUDIBLE]

LARRY KLEINKEMPER: What country are you from?

AUDIENCE: [INAUDIBLE] San Antonio. Yes.

[LAUGHTER]

AUDIENCE: [INAUDIBLE].

LARRY KLEINKEMPER: Woo, Texas! OK. We'll get to you. You just wait. Wait your turn. We're going to take care of the guests first. And then we'll get around y'all. OK? So where-- yes, ma'am?

AUDIENCE: Sweden.

LARRY KLEINKEMPER: From Sweden. Yes. Well, welcome. Trust me, we're all extremely excited to have you here. So very cool. Who came the furthest in the United States? We got anybody from way out in Maine or someplace?

AUDIENCE: Florida.

LARRY KLEINKEMPER: Florida.

AUDIENCE: Rhode Island.

LARRY KLEINKEMPER: Rhode Island's represented. Half his state is here.

[LAUGHTER]

Is that right? I think that's the population, roughly. And then who else is here?

AUDIENCE: Texas.

LARRY KLEINKEMPER: Texas. Craziest thing happened last week. I went to the Quad AIA Convention in New York. And it was New York, New Jersey, Pennsylvania, and Connecticut. All AIA went to this conference. It was a whole 400 people. If you go to the Texas AIA-- who went to Texas AIA last week? Anybody? Yeah. It's like 4,000 people for the one state. It was just-- it's crazy.

So anyways, can I begin this thing yet? Or should we? Because I mean, it's either that, or I can sing "The Eyes of Texas" for you. And no, we should go ahead and go and begin.

All right. Well, welcome. Howdy! Hello! All right. Let's get this show on the road. So we're going to talk about reality capture. The city of San Antonio has this place, this church. OK? This what used to be a church. And it's called the Alamo. And people come from all over to see it. And it recently became a UNESCO heritage listed site. World Heritage site.

And they had one little problem with the Alamo. People would walk up to it and say, [GASP]-- that's it? And there was a lot of misconceptions about it. People would think that the courtyard in the back was where the fort was. And it just really wasn't-- it was having a hard time telling the story of what actually happened there. And it is a really important part of our history.

So they were going to do the master plan to revitalize that area, and make it more contextual, and help people understand what's going on there. OK. Where do I point? Oh, maybe I should stop hitting the Back button. Oh by the way, I'm Laser Larry. Hi, everybody! I should have started with that.

So anyways, I'm an American tech who plays with lasers a lot. And my role in life is to help further the profession through technology. You know, I'm not a great designer. But I do that So let's have some fun together.

Right. So reality capture. This is a term that's being thrown around a lot. And it means a great many things. It's a very broad and generic discussion point. And when you say that I'm going to do some reality capture, you could be talking about a number of things.

So we're going to focus on some of the ideas like digital cameras, 360 cameras, video cameras, drones, terrestrial scanners, mobile scanners, and then distance meters. Who put that in there? That's not right at all. OK.

So anyways, I think what they meant was a DISTO. So-- oh. I pressed the button like I was working a DISTO. Sorry. And basically, an electronic tape measure. Right. So if you talk about reality capture, the words kind of look like this. It's all over the map.

The important thing to know is that every tool has pros and cons. And how do you pick the best tool for the job? So our task was to create a computer model of, for a large part, San Antonio. So the historic San Antonio documentation case study. And this is a mixed reality-- mixed media in reality capture.

So the area that they wanted-- and I got to make sure I press right button here-- was this area of downtown. Was really important to them. And that was about 1.5 square miles. For those of you in the US. For those of you working anywhere else in the rest of the world, we're talking about 3 square kilometers. Just to help you out.

And four weeks of data collection, running around scanning and gathering. Six weeks for modeling that area of all the buildings. And then we needed a Revit model that pulled together all the data sources. And then we took that model. For documentation, they used Revit. For rendering and animation, they used-- welcome. Come on in.

Sorry, I didn't realize that they had not let the late comers-- or not late comers. Unregistered. The cool crowd in. They hadn't let the guys that don't sign up for anything. Come on in. Come on in! We're glad to have you here. Is that the rest of the line? There's plenty of seats. We're just in the middle of having fun. So welcome. Hello. I'm Laser Larry. OK, you've caught up.

OK. That's pretty much it, right? So anyways. OK. So we're talking about reality capture. And we're talking about how to pick the right tool when you've got to do a gigantic area. And this was an urban environment.

So we broke the model down into three real areas of concern. So in white here, that was area one. And that was super intensely important. That is the Alamo and the Alamo grounds, and very directly adjacent.

And then there was a second area that may be hard for you to see that's outlined in blue. And that's the area that's directly adjacent. And it's pretty important And then area three in the dark red, that sort of background, and the environment in which this project lives for a large part.

So if you think of a painter trying to do this, you can imagine the idea that this is fine detail, high level of detail. You might use a bigger brush to get it done quicker. And then this is the very broad brush that you're slapping it on just give background. Right? Same sort of concept.

So Langan was a head partner on this project. We've got Matt here, in case your questions get very technical in a bit. And they did all the 3D scanning. They did mobile scanning and terrestrial scanning, which I'll explain what those are in the moment.

And then we also used aerial photography from PWL Model Works. I said photography, I meant photogrammetry. So photogrammetry, we're going to explain what that is as well, and explain how it's used.

So if I can position this, I might be able to stay focused. Yes. So zone 1. Zone 1 was the most critical area. That's where the Alamo is. That's where all the buildings on the Alamo grounds are. Extremely important to preserve and protect.

So for that, in order to get the highest level of resolution of the model, we were using terrestrial scanning. You can see there we got P40 out. Whoops. Sorry. They put the laser beam too close to the Next button for me.

So we've got a P40 out there, and we've got a person from Langan running it. And they're going all around this thing scanning it in high resolution to give us the best 3D measurement of the building, and the most accurate information we can get. And they're tying that to survey control. Yeah. We use P40, P20, and a C10 on that. And yes, I mentioned we tied it to control.

OK. Zone 1 terrestrial. The reason we did that. The benefits of using that scanner on a tripod is high accuracy. The manufacturers downstairs, they'll talk to you about, oh, it's within 2 mil. By the time you tie all the scans together and registration, I would say it's closer to-- I like to promise my clients locally, it's going to be a quarter inch. And if we do better than that, rock on.

So for those of you from the US, when they talk 2 mil, that's about an eighth of an inch. And a 1/4 inch, that'd be about 4 mil. 4 to 5 mil Is what I'm thinking.

High data. You get really dense data. It's also very heavy in the computer. We're talking about terabytes of data. Hundreds of gigs at the very least. And we did color scan it. Although honestly, color scan data, I tend to only recommended for plant rooms that have color coded pipes, and things that you might want to use for marketing. We might use this for marketing at some point. Especially in Texas.

And then, but the drawbacks. It does take much longer than some of the other techniques we're going to talk about. So if you have to do a whole city, this may not be the right tool for it. This is the right tool for buildings, and getting in close, and getting a high level of accuracy, but not doing large swaths of area unless you've got a lot of time.

Huge data sets. We talked about that. A need to empty area. So this was a requirement not of the technology, but of the client. And also the fact that this is such a crowded place.

So it is eye-safe. We try to keep people at least a meter away, is what they say. But honestly, that's more to keep them from kicking the tripod over for that $120,000 instrument. And so this is what's important to us.

You can scan in a room. For example, if I set it up right here, can I still figure out where the walls are and how far apart they are? Can I still see the floor and the ceilings? Yes. Are there some shadows? If I need to detect where's the electrical outlets back there, then I might be more concerned about whether or not there's people in the room.

But there was also a privacy concern from the Alamo. And so they had decided we're going to wait till at night, and we're not going to scan it when there's people so that we can capture everything.

And then why terrestrial scanning? Highly historic area. Data will be generated in the future. And also areas of potential for new construction. Wrong button. Right. I'm pressing the laser again. Do it. You can do it. You're a winner.

OK. So that's what this looks like. And so you end up getting-- this is color scanning. You'll see things like, this is actually a little bit of a shadow where you didn't get on top. We didn't get necessarily on top. The color is not amazing. But we've got ways of making that better.

But this is right out of the box. Once you register it together, the different scanning setup is what you get. And then you can build a computer model off of that. And you can compare it over time. One of the things that we're looking at is the idea of comparing scans before and after construction and as time goes on, to see how the building is decaying.

Right. So ortho-photography. I've been complaining or talking a little bit about the color in the scanning and the way that ends up working out. So for where we were going to do high resolution rendering and animation for marketing pieces and for the master pieces, we went through and we took ortho-photography.

The benefits. You get HDR. You can create great texture maps for the computer model. Drawbacks. It's a very time-consuming process. High level of manual labor. And also, we only did it in the historic areas where texture maps would need to be for the future models.

And that looks like this. So you get great color. You can balance it out and everything. And that ends up being wonderful texture maps for the building. It makes it look real almost photoreal. Whoops. Oh, yeah. That one. Just a little more detail on the inside. So flattening that out so that you can see all the stones and everything, and where it's been patched. You pick up a lot of detail.

Also we we're scanning at night. Langan was scanning at night. And so a lot of the interior spaces, or around it, may not have been the best lighting. So HDR was super important for what we wanted to do in the end.

Right. Area 2, directly around the Alamo grounds, and out, say, a block in every direction. We did mobile scanning. So Langan took their Pegasus scanner. And in some areas, they put it on a cart. Other areas, it's mounted to the back of one of their trucks. They ran it around, and drove through the streets.

Additional sensors. They also had use GPS. Yep. And they were tying everything to a control network so that it would mesh with their terrestrial scanning that they had done. So you're combining mobile with terrestrial.

The benefits are it's really fast. You're driving through and collecting stuff. The area can be occupied. You don't have to, other than waiting for a traffic light and people to cross the road, you just drive through and you get them.

Drawbacks. It says lower accuracy up there. I want to explain that a little bit. So it's pretty darn close to the accuracy that you're going to get with terrestrial scanner per measurement. The difference is, if I was measuring this screen, terrestrial scanner is going to give me two dots right here. Right next to each other. And this whole thing would be covered with dots. Right?

So with the mobile scanner, I'm going to get a dot here. And then depending on the resolution we set it for and how close it is to the mobile scanner, I may have another dot here. Or if it's further out or a lower resolution, it may be a meter away. So somewhere between or 300 mil to a meter is a typical range of settings. And also, the further you get away from the scanner, the more the dots may spread out for the measurements. So we consider that a little bit less accurate as far as the amount of detail you're going to pick up.

Drawbacks. And also what I hadn't mentioned with the terrestrial scanners, we get these 360 degree bubble views, much like you see on Google Earth on Street View. So we can set up a plan that has all of those. And you can click on them and you can see everything, all the way around. I got a little dizzy there. That's part of dehydration. But on a mobile scan, you can't do that.

So that's a little bit of a drawback. I found that a lot of clients who don't, they love that street view of their building with the scans. And they don't have to have a strong computer to use it. And so a very useful tool.

Also on the buildings for master planning, we had the lowest resolution that we normally would do for a building, which is 6 inches. If it's smaller than a 6 inch detail, we really don't need it. We're just trying to express what's going to happen in the planned area. OK.

And the mobile scanning creates something like this. So there now, you can see, the Alamo, all the stuff that we did with terrestrial scanning, is going to fit nicely in there. And then this gives us all the measurements on the background buildings that we're going to need, and the sights, and tells us what the terrain is doing.

And then the last big part of the equation is for the buildings further out. So we're in the middle of a city that has a lot of skyscrapers. A lot of tall buildings. And how what we do is going to affect those view quarters. And if there's anybody in here from England, they have a lot of regulation on right to cite, and right to light, those sort of things.

And so San Antonio wanted to be very sensitive to that as well. And so they wanted to be able to see what they were doing and how it was going to affect the views. So instead of running around scanning and modeling at a high level of detail all those skyscrapers, we opted for photogrammetry from aerial photography.

So a lot of people are doing it with drones. This particular case, we had an airplane with a high resolution camera. Takes a lot of photos. And then we put those photos into the computer, and it pumps out a point cloud. And then the good people Pictometry, they take that as well, and end up meshing the model. They end up making a surface model with all the textures mapped to it, much like what you see on Google Earth. But this is one that we have that we can tear apart and rebuild to insert our high level of detail model in the middle.

Benefits. Compared to the rest of the model, it's the lowest cost. High resolution aerial photography looks great, and can be purchased. It can be textured through the photogrammetry process.

Drawbacks. It is the least accurate of all the techniques. And so the way I like to describe it, when it comes to using that technique. If you're doing campus planning, fantastic. It's the cheapest, easiest way you're going to get it. But if you're placing steel, hell no. OK? Do not-- no! You don't do that. OK? Right. So that's just-- I'm probably not supposed to cuss in these sort of things.

Anyways. So why photogrammetry? These were only areas that were considered background environment. And so low level of accuracy didn't bother us. And predominantly just a need for sense of scale. Make sense to everybody? Everybody following what I'm saying? I'm not talking over anybody? OK, good. I think I'm keeping it at a reasonable entry level, sort of. And we'll get more technical in a little bit if you want to.

So what do you get? You get something that looks like this. And we had already subtracted out in this situation our area. There's where the main area of focus is, and some of the mobile scanning has been deleted out.

All right. But if you get up close. So it looks-- and I'm going to go back-- looks pretty nice, doesn't it? Yeah, it 's like Google Earth. Nobody's impressed anymore. OK, that's fine. But if you get up close-- OK, you think I wasn't impressed before? Now I'm really not impressed.

So up close-- crap. There's holes in the terrain. Things, textures are you just sliding off of them. But you'll notice the skyscrapers in the background further away from where you're standing-- OK, that's acceptable. I might need to fix a little texture map right here. Or that's the church steeple. That's-- oh, I just realized that's a church steeple in this image.

So there's no problem with that texture map back there. Tower looks good. All that, I think, most of that's acceptable. I might play with the texture's a little bit. But for the most part, we're good to go. OK? But this area. Yeah, this area's crap. We got to do something about that. And that's where our good friends at Langan and us worked together to replace all of that.

So in the Revit model for the documentation side goes, the model looks like this. The textures unfortunately do not transfer into Revit. We have them in 3D Studio Max for doing the rendering animation in real time. But Revit-side, those textures don't transfer back that way.

Do we have anybody from the Revit development team in the room? Yeah, you jerks. OK. [LAUGHTER] They have more work to do. So anyways. Cool program. Totally cool program. Love it.

So anyways. But, yeah. That's where that's at. OK? And it looks like this. So we've got a sense here. We don't have to make them the blue. We change that color, or whatever. But they're massing. They're background. Gives you a good sense of scale.

All right. So again, just to drive it home. So the green lined area, this was all terrestrial-scanned. This was modeled real pretty. The yellow area, we did these buildings back in here with mobile scanning. And then the buildings way in the back, we did with the aerial photography photogrammetry. And then combined them all together in order to give us something that had the right amount of accuracy in the right timeline in order to build the whole city in six weeks, but had all the information that we needed.

Whoops. Sorry. And in the Revit models, they ended up looking like this. So this is in the high to medium area. Oh there's, Veronica. Hi, Veronica! Everybody say hi to Veronica.

[AUDIENCE QUIETLY GREETING VERONICA]

So anyways. Veronica was job captain on this project for the Revit side. And then she passed the model over to me, and I took it over to the Max side. And so a very cool person back there. You all want to get to know her.

So anyways, Veronica, was this is in the high level or the medium level of modeling? It was the 6 inch. OK, so the medium level. Right. OK. So--

AUDIENCE: [INAUDIBLE]

LARRY KLEINKEMPER: Whoops. You were having too much fun, weren't you? That's one of the hardest parts my job, is I have to come through, and I'll be like, you're having too much fun. They're done. I'm losing money. Keep going! [LAUGHTER]

So anyways. But they enjoy their job. OK. So lessons learned. We need to distinguish between needs and wants for a project in order to optimize your resources. Very important.

Manage the expectations of the client. A lot of clients, they don't know. They assume everything's really expensive, and that it's a push of a button. And so you've got to try to help them through what you're going to get out of each part.

And then have a modeling protocol. A project like this takes a team. It's not just Veronica. Veronica was job captain over, like, six people on this project. And so it's important for everybody on the team to know this area we're going to hold to 6 inches, and not model anything smaller. And this area, we're going to try to get all the detail in it we can. And this area, we don't even care about. So don't bother with that.

And then, have a plan for dealing with the boundary condense-- [BABBLING]-- the boundary conditions. So where one zone meets the next zone which meets the next zone, you have to put some time in for merging those two and blending the transition so that it's not a hard transition between beautiful model and, this is crap! We want it to blend over and then, like, ah, you didn't see it. The crap's way too far away. I can't see it. Right.

Hey, I'm Laser Larry! And that is reality capture and using mixed media in order to create a downtown area. I think I went fast enough that we have time to get into the technical questions. How close are we to 3:15? Anybody with a--

[INTERPOSING VOICES]

I got half an hour. Wow, I went really-- I cut out slides because I tend to embellish. And a lot of times, I take too long. So, OK. Technical questions. Who's got some. Yeah, go for it.

AUDIENCE: So you captured this area.

LARRY KLEINKEMPER: Yeah.

AUDIENCE: Then you leave out areas for each individual [INAUDIBLE], so you can texture it and make it more appeasing.

LARRY KLEINKEMPER: That's correct.

AUDIENCE: And then you put everything georeferenced back into what--

LARRY KLEINKEMPER: So it was all, except for the photogrammetry, it was all georeferenced to begin with. Thank you, Mr. Thank you, Matt at Langan. So they did a beautiful job with the surveying. And so that was already there.

And then we don't ever move it. This is, if you get into 3D scanning and turning it into Revit models, you don't want to move their point clouds. Ever. You want to build on top of that, because if for any reason you were to find a problem, which we did not find any problems with your stuff. But with a lot of people, we do.

And if they have to go back and fix that, it will not pop into the exact same spot if you've moved it Go figure! And that causes problems. So if you have got a great surveyor on your team, then and it's one and done, fantastic. But not everybody is one and done. And some people-- well, just some people are better than others. It's the nicest way I can put it.

AUDIENCE: So was there any issue with the texturing because the local systems work better with texturing? You're that far out. Is there any issue in terms of the material you were using?

LARRY KLEINKEMPER: Right. Great question. So as far as coordinate systems go, luckily, we had them place the coordinate system so that the origin was very close to the model. Close to the center of the Alamo. That is an incredibly important point that you're making, that everybody is just like, why is he concerned with it? Super important.

OK. So most surveyors in this world love-- sorry, not this world. Is it this world, or just the United States?

AUDIENCE: I'd say it's this world.

LARRY KLEINKEMPER: OK. They love this thing called state plane coordinate systems. Which means the origin for your model is, like, 1,000 miles away. And if you ask the Revit development team-- they're still not here, right? There's nobody from the Revit develop-- OK.

So they're going to tell you that, uh, our program doesn't really like to go over 20 miles, if you don't mind. After that, it kind of becomes a little inaccurate. Because it only rounds to, like, the fourth decimal point or something.

In fact, when they say that, they mean 20 mile diameter. When in fact, from the origin, you can only get about 10 miles before you start seeing this phenomena. 3D Studio Max also has this exact same problem. Further away from the origin you get, suddenly you try to pick on something. And as you zoom in, it's like, I'm not here! And you try to pick on it, and it goes, no, I'm over here! OK, I'll pick up-- now I'm over here! You know?

And it's like, oh! You're annoying! OK? It's really-- sorry. It's one of the banes of my existence. So when you do the 3D scanning of a very large area-- and I'm so glad you brought this subject up, because it is super important-- you have to remember to ask your very well trained surveyor to use a coordinated system, or to copy the coordinate system so that the origin is in the center of the model, or very close to it-- within 10 miles.

If you do that, which they did for us-- thank you very much-- what that does is then you don't run into those accuracy issues that you were just discussing. And it makes your whole life better. Anything to add on that, Veronica? Did you? OK. All right. Matt, do you have anything? OK. So anyways. Yeah, please.

AUDIENCE: [INAUDIBLE] georeferencing and [INAUDIBLE]. [INAUDIBLE]?

LARRY KLEINKEMPER: OK. I'm going to repeat your question to Matt so that he can answer that. So what this very smart gentleman was asking was OK, you've done this great survey with terrestrial scanning. How do you tie that into the survey control for the mobile scanning?

MATT: So we do that always the old-fashioned way. We ran a total station down on the streets. And we actually set Google Maps. We had set Chevrons all down the street [INAUDIBLE] using the total station. [INAUDIBLE].

LARRY KLEINKEMPER: Yeah, I didn't mention that.

AUDIENCE: [INAUDIBLE] photogrammetry?

MATT: The photogrammetry actually wasn't-- we didn't apply the photogrammetry.

LARRY KLEINKEMPER: Right. So the photogrammetry, again, least accurate. It basically was fit. So in fact, photogrammetry, out of the box, totally unscaled. The entire city, I believe, was a little over 4 feet wide. OK.

AUDIENCE: [INAUDIBLE]?

LARRY KLEINKEMPER: What's that?

AUDIENCE: [INAUDIBLE] within 6 inches?

LARRY KLEINKEMPER: Within 6 inches? OK. But what we did was we used his mobile scanning as our reference point, and scaled the model to match that. And for background buildings that we weren't placing steel or anything like that, it was just for a visual reference, that was accurate enough in that situation. Great questions. Yes, sir.

AUDIENCE: What method did you use to create the orthographic images that you applied those texture maps on?

LARRY KLEINKEMPER: Oh, yeah. So basically, high resolution cameras. Then you have to go through Photoshop and flatten those things out to get them to be-- we want it to be as straight on as possible. And then in 3D Studio Max we're applying those textures to the model. And typically, it's just like a UVW WMAP map. We're not doing-- we do unwrapped sometimes, but that's really kind of overkill for us most of the time. Lady in the back.

AUDIENCE: I had a question just about the process with you fitting point clouds from your scanner and then inputting that in Revit. Is it something where the software prefers the mold or the mass, or is that where you're actually creating a mass based on what you see in the point cloud? Like using it as the actual model?

LARRY KLEINKEMPER: OK. Wonderful. So you think you've asked one question. When in fact, I have to give you five answers. So here we go. Are you ready? Everybody stretch out a little bit. Here we go!

Matt creates a point cloud. He goes out, he scans. He runs his little cart around. And then he registers his point clouds together. And that's where he hands that off to the architects and the engineers. Or Landmark Services, if you prefer. We'll be happy to do it for you.

You bring that into ReCap if you're working with an Autodesk product. And seeing as we're at Autodesk University, I'm assuming that you're not all working with Bently, and you're working with Autodesk products.

ReCap turns it into a point cloud type, or file type, that all-- not all-- Revit, AutoCAD, 3D Studio Max can read. I'm unclear as to whether or not Inventor or Maya can bring it in. And Navisworks. Thank you. You're absolutely right. Thank you. I forgot that one.

So then, in those programs, you import-- and each program does it its own little way. Because they're one company, but like every program, I'm my own little program! Can I have my allowance? So anyways.

[LAUGHTER]

Like, not everything-- we can't standardize this. So each one of the programs has their own way of importing a point cloud. And you have to look up how to do that. Now, once you do that, then becomes the process of understanding. I like to call Veronica and I the radiologists of architects. We're the people-- if you've ever had a baby or seen one of those ultrasounds, you hold it up, you can kind of see a baby.

And then the nice lady who runs the ultrasound always says, it's a boy! And you're like, it's static!

[LAUGHTER]

I have no idea what the hell I'm looking at here. You know? And you're like, oh, wow! It's huge! No, that's his head. What I'm talking about's over here. Right. So when you zoom out, it's super easy to see it's a building. And you can clearly see that you're building.

But when you zoom in or you try to get into the interior spaces, it can confuse people often. It can turn to dust, and just look like a series of points. And people can get kind of lost in it. That's all we do, is we're doing scan to BIM every single day, modeling from point clouds. And so we're like the radiologists who can go, it's a boy!

And, yeah. But over time, it gets easier to use, sort of thing. Anybody else?

AUDIENCE: [INAUDIBLE]?

LARRY KLEINKEMPER: It was gigantic!

AUDIENCE: Like, is one structure a bunch of meshes or bases, or--

LARRY KLEINKEMPER: Oh, no. OK You're right. So, OK. So let me backtrack. So again, you asked one question, and here comes the answers. And it's multiple. So Matt hands over a point cloud. Your surveyor, or whoever runs your 3D scanner, runs-- by the way, I am not a surveyor. I will not be a surveyor. But I will tell you, and people who own scanners hire me. And sometimes I hire them.

A surveyor will run that scanner better than anybody else. They've definitely got that system down of trying to run control and adding control. That surveying background is super important. If you're not a surveyor, maybe take a surveying class. Or something like that to help you out. Most people do good, they do better.

Anyways. So the point clouds, they're gigantic. For a project like this, we're talking no less than 500 gigs. And no more than maybe 5 terabytes. So a lot of people's computers and systems have trouble even loading or running them.

We then boil it down into-- we boil it down into a normal Revit model. And that brings the file size down to maybe 100 megs for something like this. And then the 3D Studio Max model, when we added all the trees and and bushes and stuff, and all the people, it probably grew to somewhere between 1 to 2 gigs for all the files and stuff. The actual Max model was a small part of that. But that's what it was.

So those were your answers to your one question. Did I answer it right? Or clear?

AUDIENCE: I was trying to figure out if each structure was [INAUDIBLE] or a bunch of [INAUDIBLE].

LARRY KLEINKEMPER: No, I mean a Revit model for the-- OK. Yeah, I know where you're--

AUDIENCE: Are you talking about the part that he did with the photogrammetry?

AUDIENCE: Yeah, like each one--

AUDIENCE: Actually [INAUDIBLE].

LARRY KLEINKEMPER: OK. Yeah, that's what I was just about to say.

AUDIENCE: Yeah. And so one is [INAUDIBLE].

LARRY KLEINKEMPER: Zone 1 was a high level, or a decent level, Revit model . Zone 2 was less resolution. Only exterior. We're only modeling the shells. But zone 3 was planars. It was planar boxes, and with texture maps.

AUDIENCE: So you could explode them and edit them, and [INAUDIBLE].

LARRY KLEINKEMPER: Yeah.

AUDIENCE: [INAUDIBLE].

LARRY KLEINKEMPER: Who else? Yes, ma'am. From San Antonio.

AUDIENCE: Where can you see the final product?

LARRY KLEINKEMPER: What's that?

AUDIENCE: Where can you see the final product?

LARRY KLEINKEMPER: So I have a number of beautiful renderings, which you can find online. But I'm not allowed to show it at a conference because the city of San Antonio and the state of Texas are still in discussions about what's going to happen.

But you can look up to see, if you just google the Alamo master plan, you'll probably find some images from the local news and other people. But I'm not allowed show them to you here. And that's really disappointing.

Yeah. So anyways. But you'll see videos and animation-- or renderings, anyways-- of all of that if you google that. Yes, sir.

AUDIENCE: I just have a quick question.

LARRY KLEINKEMPER: Yes, please.

AUDIENCE: So you did scan the inside of the Alamo too, right?

LARRY KLEINKEMPER: Yes, we did.

AUDIENCE: You mentioned you were just doing shells based [INAUDIBLE].

LARRY KLEINKEMPER: No, no. The Alamo itself, the church itself, is then scanned ad nauseum. And for a really good reason. The Alamo is actually disappearing in front of our face. The stone that the Alamo is made of is decaying. And it's actually, dust is falling off. Particles of it is are falling off. It is deteriorating in front of us.

And so one of the things that we're really interested in discovering as we go through this process is, how fast is the rate of deterioration? And so we plan on scanning and rescanning and scanning it again over time to start to show where it's decaying, and try to understand why it's decaying in those areas faster in some areas than others.

It's been repaired over time. It's had different types of mortar applied to it. Lots of things have happened to that building over the time.

One of the most interesting things that I had learned was the idea that when you walk into the Alamo, you're actually 2 feet above where the surface of the church was when it was originally built. So if you can imagine-- and so is the outside. The inside and the outside.

So if you can imagine, the Alamo is actually, over history, you would believe that it has shrunk by 2 feet. But what's happened is the ground has actually come up. Over years, they would apply more and more surfaces. So part of the master plan process that I think this part, everybody agrees on, is that they're going to remove and get back to that living surface of where it originally was.

Everything else, total argument. Like there's battles still going on at the Alamo. But that part. Yeah.

AUDIENCE: When are you guys planning to do your scan for the second half?

LARRY KLEINKEMPER: That has not been determined.

AUDIENCE: So the timeline is still being discussed?

LARRY KLEINKEMPER: Yeah. That's part of the monitoring process. And that's all being still worked out and planned out.

AUDIENCE: When was your guys' scan completed?

LARRY KLEINKEMPER: A year ago.

AUDIENCE: Yeah.

LARRY KLEINKEMPER: Yeah. Yeah, this was, we started this-- I know we were just receiving their scans about the time that I was at AU last year.

AUDIENCE: You guys scan that stuff first [INAUDIBLE].

LARRY KLEINKEMPER: Yeah. Anybody else? OK. Well, listen. Take a sort of short class, and be grateful I didn't bore you to death.

[LAUGHTER]

[APPLAUSE]

Thanks, everybody.