This brings back memories of Microsoft's acquisition of SeaDragon. At the time they had a really compelling demo (at least for me) of reconstructing 3D locations based on a smatterings of photos.
Seeing a digital version of it in such detail only further reinforces how important it is to experience it in person.
Few sights of man-made things have instilled as much awe in me as La Basilica Di San Pietro and most of them are also in Rome (namely the Pantheon and Moses @ Basilica di San Pietro in Vincoli).
I was there two weeks ago. The tour guide took us through a route that bypassed the longer lines and through some underground areas—culminating in an entrance that completely blew my mind. I never realized how huge the interior was until I stepped in and saw it firsthand. There are few things in my life that completely took my breath away, this ranks in the top 5 for sure.
Exactly. And it was a great marketing tool for catholicism, imagine simpler (even if rich) folks came to visit the pope and experienced this marvel of medieval construction. You feel utterly insignificant on purpose, feeling weak and in presence of something much larger is an easy way to more faith, a truth valid for all humans across all time.
But to me, despite all of this, there was a lot of sadness in that experience - because you know how desperately poor common folks were, how instead of building such status mega symbol they could have done some proper good. But not for church of that era, it was busy fighting for power and money of that world and trying to show how above everybody else they were.
You can see miniature scale of this in literally every (also non-) older European village or town - religious buildings have received by far the most funding and care, sometimes overshadowing kings castles themselves. Cathedrals were always built to impress masses, and this one is just on top of the game, by huge margin for good reasons I believe.
Interesting that people may experience it differently, but to me it was a bit of a letdown, somehow it felt larger than the human scale, so maybe impressive as a technical feat, but also somewhat boring, more intimidating than moving -- I was more touched by some of the others in Rome that you mention. But to me the ultimate awe-inspiring church was the Basilica of Assisi that felt just perfect in proportion and design.
Yes. Impressive tech but the website is ultimately not a great experience. You don't get the detail, the texture, the light, the human scale etc. Instead you get bits of wire frame, stuttering, odd flying movements, anti-aliasing issues etc. And a forced narrative along the side.
Agreed. St. Pete’s is so huge and perfectly proportioned that it feels not of this world. To think they fumbled around for generations until Michelangelo took over tells you that he was an every thousand years kind of dude.
And oh my, the Pantheon. My wife saved it for a surprise as the family walked through Rome. We approached it from the back. To me it was just another dirty old building from that angle. She hustled me through the entrance while I was keeping track of the kids so I had no clue. I… I may have wept.
I have a picture of the light streaming down through the Pantheon onto one of the sculptures. It's the best photo I have ever taken and every time I look at it I remember that feeling I had looking up at the dome. Really a special thing to experience.
Not an explicit field expert, but pretty well into computer graphics(games) and been reading a bunch of papers in the field over the years.
Classic photogrammetry was always a mixed bag in terms of results (especially if trying to construct meshes), but even before NeRFs (Neutral Radiance Fields) and Gaussian Splatting there was a ton of work using neural nets to handle various parts and I doubt that many modern tools avoid using them.
So in a way, these fields actually made use of neutral nets/"AI" (honestly more relevant imho than most of the LLM stuff).
I think putting AI front and center in the marketing like this is a public relations move by Microsoft to brush up the image of AI in the general public.
It's not preserved until the data and source code are open, I'm sure these corporate exercises are impressive to potential clients, but they have absolutely nothing to do with preserving, studying, or expanding access to art and culture.
Yup, this is culture-washing financed with MS dollars, who finally found one way of promoting AI for Good™ (literally spelled out as such in the marketing video), probably hoping it would make up for the Gen AI slop humanity will have to deal with for decades to come.
dumb question: could we, in the future, use some kind of gen ai to generate a videogame map (i'm thinking quake 3 arena / openarena) of buildings like these ?
I think the biggest challenge is not any of the technical or legal problems already mentioned, but that none of these buildings are laid out with the primary objective of being fun to run around shooting people in. So once the novelty wears off, I expect the actual gameplay experience will be rather clunky, especially with competitive gamers.
With players in control any jank will be quite obvious, the field did accelerate thanks to neural nets but there seems to have been a lot of focus on NeRFs and GS (This interactive demo seems to use GS) and classic triangle-geometry (especially lower polygon counts) hasn't gotten as much love recently as the impressive GS demos has taken over.
But the success of GS and speeding up should rekindle some interest and let us use some of the advances in making "production ready" methods.
Whether you can make reproductions of buildings and public interiors is known as "freedom of panorama". Wikimedia Commons has a comprehensive list by country [1].
I wouldn't care about reproducing existing building as long as the AI can generate credible ones in the same style, then place them on dynamic worlds created by prompting the AI. Having intelligent AI NPCs as long as AI generated scenery would be a killer feature in any game. I'm talking about off line disconnected single player games; cloud ones with these features could be already here, but I want to be in control, and marketing rules are against that: who would buy the new shiny V2.0 with the new worlds and characters if 1.0 could create them just by asking it to?
As someone who consumed tons of scifi novels and books as a kid and wants to be immersed in big worlds, enjoying great stories also in games (absolutely loved the Mass Effect saga), I already know what's going to happen when we'll be able to feed Philip K. Dick or Asimov, Sturgeon, Bova, Silverberg, etc. books to an AI and have it create worlds, environments, stories and characters straight out of the book descriptions. Literally drooling over it.
It's like that kid that got expelled for creating a map of his school in Counter-Strike [1], due to fears of security threats. Not that I blame them, I could see people planning a robbery in Minecraft.
To further sustain this point: I heard that in the past, someone recreated some parts of Politecnico di Milano (a famous technical university in Italy) as a map of some open source first person shooter. Unfortunately I don't remember which shooter it was.
One such pipeline that already works today is photogrammetry of real place -> voxel data using VoxelPlugin. You can then leave it as a Voxel or bake it to a static mesh.
A lot of advances during the Renaissance happened due to some vanity projects of the Medici family (e.g. funding Galileo Galilei and Brunelleschi's Dome).
I sometimes wonder why they don't just sell the whole thing to a real estate developer or something.
If inornate churches that look more like strip malls and expo centers are so much better for the laity then imagine how much more good it would do for the top brass, relieving them of the burden of having to look at all that sacred art all day long. What's good for the goose is good for the gander.
This brings back memories of Microsoft's acquisition of SeaDragon. At the time they had a really compelling demo (at least for me) of reconstructing 3D locations based on a smatterings of photos.
https://en.wikipedia.org/wiki/Seadragon_Software
https://www.youtube.com/watch?v=IFSsTwXLqsc
Thanks for the memory - definitely one of the coolest demos I'd ever seen.
Any idea how accurate the pitch was compared to the reality?
At the time I really thought that was going to be the next Street View.
Maybe with WebGL and Gaussian Splats that can still be the case. But it’s also a ZIRP kind of project. Awesome, but what’s the business model?
Seeing a digital version of it in such detail only further reinforces how important it is to experience it in person.
Few sights of man-made things have instilled as much awe in me as La Basilica Di San Pietro and most of them are also in Rome (namely the Pantheon and Moses @ Basilica di San Pietro in Vincoli).
You can't understand the scale of it until you experience it in person. The way I thought of it was that it is a cathedral made for giants.
Totally agree.
I was there two weeks ago. The tour guide took us through a route that bypassed the longer lines and through some underground areas—culminating in an entrance that completely blew my mind. I never realized how huge the interior was until I stepped in and saw it firsthand. There are few things in my life that completely took my breath away, this ranks in the top 5 for sure.
It's also a symbol of all the money and gold, the real values and the secrets of the church...
If God exists, you think he would want you to sacrifice and spend it all on gold and salaries of locals ?
1 reply →
Exactly. And it was a great marketing tool for catholicism, imagine simpler (even if rich) folks came to visit the pope and experienced this marvel of medieval construction. You feel utterly insignificant on purpose, feeling weak and in presence of something much larger is an easy way to more faith, a truth valid for all humans across all time.
But to me, despite all of this, there was a lot of sadness in that experience - because you know how desperately poor common folks were, how instead of building such status mega symbol they could have done some proper good. But not for church of that era, it was busy fighting for power and money of that world and trying to show how above everybody else they were.
You can see miniature scale of this in literally every (also non-) older European village or town - religious buildings have received by far the most funding and care, sometimes overshadowing kings castles themselves. Cathedrals were always built to impress masses, and this one is just on top of the game, by huge margin for good reasons I believe.
9 replies →
Interesting that people may experience it differently, but to me it was a bit of a letdown, somehow it felt larger than the human scale, so maybe impressive as a technical feat, but also somewhat boring, more intimidating than moving -- I was more touched by some of the others in Rome that you mention. But to me the ultimate awe-inspiring church was the Basilica of Assisi that felt just perfect in proportion and design.
Yes. Impressive tech but the website is ultimately not a great experience. You don't get the detail, the texture, the light, the human scale etc. Instead you get bits of wire frame, stuttering, odd flying movements, anti-aliasing issues etc. And a forced narrative along the side.
Agreed. St. Pete’s is so huge and perfectly proportioned that it feels not of this world. To think they fumbled around for generations until Michelangelo took over tells you that he was an every thousand years kind of dude.
And oh my, the Pantheon. My wife saved it for a surprise as the family walked through Rome. We approached it from the back. To me it was just another dirty old building from that angle. She hustled me through the entrance while I was keeping track of the kids so I had no clue. I… I may have wept.
I have a picture of the light streaming down through the Pantheon onto one of the sculptures. It's the best photo I have ever taken and every time I look at it I remember that feeling I had looking up at the dome. Really a special thing to experience.
This reminds me strongly of Microsoft Photosynth. Can't help wondering what the lineage between the two looks like.
https://en.wikipedia.org/wiki/Photosynth
https://medium.com/@dddexperiments/why-i-preserved-photosynt...
Yes! I forgot about this but I knew I've seen something very similar long ago also by Microsoft. I wonder if any Photosynth DNA got into this.
Direct link to the virtual tour: https://virtual.basilicasanpietro.va/en
What does AI have to do with this? They thoroughly scanned the thing, where's the need for AI?
Photogrammetry has rebranded itself to "Spatial AI".
Not an explicit field expert, but pretty well into computer graphics(games) and been reading a bunch of papers in the field over the years.
Classic photogrammetry was always a mixed bag in terms of results (especially if trying to construct meshes), but even before NeRFs (Neutral Radiance Fields) and Gaussian Splatting there was a ton of work using neural nets to handle various parts and I doubt that many modern tools avoid using them.
So in a way, these fields actually made use of neutral nets/"AI" (honestly more relevant imho than most of the LLM stuff).
2 replies →
I think putting AI front and center in the marketing like this is a public relations move by Microsoft to brush up the image of AI in the general public.
Absolutely gorgeous imagery, and it seems to have a functional purpose as well, as a digital twin for structural modeling.
Incredible work.
The work in the related stories are equally gorgeous. Thanks for sharing mate.
Looking at the source code with web inspector it seems to be powered by 3D gaussian splatting and BabylonJS (https://doc.babylonjs.com/features/featuresDeepDive/mesh/gau...).
Babylon.js is funded (owned?) by Microsoft so anytime you see web 3d on a Microsoft site it's almost certainly going to be using that.
It would be nice if they'd let you see the whole thing at once.
It'd be nice if field of vision was universal and not rate limited, or limited to a narrow band of the en spectrum...
Any way to experience this on a VR headset? I have a Valve index
Chrome-based browsers only I presume :')
Works fine in Firefox for me.
It's not preserved until the data and source code are open, I'm sure these corporate exercises are impressive to potential clients, but they have absolutely nothing to do with preserving, studying, or expanding access to art and culture.
Yup, this is culture-washing financed with MS dollars, who finally found one way of promoting AI for Good™ (literally spelled out as such in the marketing video), probably hoping it would make up for the Gen AI slop humanity will have to deal with for decades to come.
dumb question: could we, in the future, use some kind of gen ai to generate a videogame map (i'm thinking quake 3 arena / openarena) of buildings like these ?
(not just the basilica di san pietro)
I think the biggest challenge is not any of the technical or legal problems already mentioned, but that none of these buildings are laid out with the primary objective of being fun to run around shooting people in. So once the novelty wears off, I expect the actual gameplay experience will be rather clunky, especially with competitive gamers.
With players in control any jank will be quite obvious, the field did accelerate thanks to neural nets but there seems to have been a lot of focus on NeRFs and GS (This interactive demo seems to use GS) and classic triangle-geometry (especially lower polygon counts) hasn't gotten as much love recently as the impressive GS demos has taken over.
But the success of GS and speeding up should rekindle some interest and let us use some of the advances in making "production ready" methods.
we could have these today; the difficult part is getting permission to use the building in your work (depending on jurisdiction / the work)
Whether you can make reproductions of buildings and public interiors is known as "freedom of panorama". Wikimedia Commons has a comprehensive list by country [1].
https://commons.wikimedia.org/wiki/Commons:Freedom_of_panora...
1 reply →
I wouldn't care about reproducing existing building as long as the AI can generate credible ones in the same style, then place them on dynamic worlds created by prompting the AI. Having intelligent AI NPCs as long as AI generated scenery would be a killer feature in any game. I'm talking about off line disconnected single player games; cloud ones with these features could be already here, but I want to be in control, and marketing rules are against that: who would buy the new shiny V2.0 with the new worlds and characters if 1.0 could create them just by asking it to? As someone who consumed tons of scifi novels and books as a kid and wants to be immersed in big worlds, enjoying great stories also in games (absolutely loved the Mass Effect saga), I already know what's going to happen when we'll be able to feed Philip K. Dick or Asimov, Sturgeon, Bova, Silverberg, etc. books to an AI and have it create worlds, environments, stories and characters straight out of the book descriptions. Literally drooling over it.
It's like that kid that got expelled for creating a map of his school in Counter-Strike [1], due to fears of security threats. Not that I blame them, I could see people planning a robbery in Minecraft.
[1] https://arstechnica.com/gaming/2007/05/student-creates-count...
To further sustain this point: I heard that in the past, someone recreated some parts of Politecnico di Milano (a famous technical university in Italy) as a map of some open source first person shooter. Unfortunately I don't remember which shooter it was.
interesting, where would one have to look to learn and/or get the necessary data to pull that off?
i might just want to do that for my own private use (or i might be okay with law infringement).
1 reply →
One such pipeline that already works today is photogrammetry of real place -> voxel data using VoxelPlugin. You can then leave it as a Voxel or bake it to a static mesh.
Example: https://twitter.com/phyronnaz/status/1549869716826689539
https://www.youtube.com/watch?v=WZbG5JTpSCA
Okay can we please play quake in this map now?
This might actually be an awesome use case.
So this is one of the vanity project Microsoft undertakes using the vast amounts of money it makes off of proprietary software?
A lot of advances during the Renaissance happened due to some vanity projects of the Medici family (e.g. funding Galileo Galilei and Brunelleschi's Dome).
You don't see any value in this project? I certainly do.
Microsoft was bummed that they couldn't acquire it so they recreated it in 3D.
I sometimes wonder why they don't just sell the whole thing to a real estate developer or something.
If inornate churches that look more like strip malls and expo centers are so much better for the laity then imagine how much more good it would do for the top brass, relieving them of the burden of having to look at all that sacred art all day long. What's good for the goose is good for the gander.
5 replies →