[HN Gopher] Map of all the buildings in the world
       ___________________________________________________________________
        
       Map of all the buildings in the world
        
       Author : dr_dshiv
       Score  : 160 points
       Date   : 2025-12-04 22:43 UTC (6 days ago)
        
 (HTM) web link (gizmodo.com)
 (TXT) w3m dump (gizmodo.com)
        
       | mrec wrote:
       | At least in Firefox, this page doesn't scroll at all. No
       | scrollbar, scroll wheel/cursor keys/PageDown all do nothing.
       | 
       | How does someone screw up CSS so completely?
        
         | dudefeliciano wrote:
         | works for me on FF
        
         | matteason wrote:
         | You have uBlock Origin or another ad blocker installed, their
         | cookie banner disables scrolling but your ad blocker is
         | blocking the cookie banner (same thing happened to me)
        
           | LukaD wrote:
           | Same here with uBlock on iOS. This happens every now and
           | then, not too often. But when it does, I usually decide the
           | page is probably not worth reading anyway.
        
         | nirewen wrote:
         | For me it was the cookie notice. Had to disable uBlock and
         | reject them before the page could be responsive again.
        
       | wongarsu wrote:
       | Github: https://github.com/zhu-xlab/GlobalBuildingAtlas
       | 
       | Viewer: https://tubvsig-so2sat-vm1.srv.mwn.de/ (might already be
       | hugged to death? When it works you get a heatmap when zoomed out,
       | and 3d models with flat roofs when you zoom in far enough)
       | 
       | Looks pretty cool. Obviously only covers above-ground buildings
       | or above-ground portions of buildings. Also seems like it tends
       | to view buildings built wall-to-wall next to each other as the
       | same building, but not always. So if you calculate something like
       | average building volume you are bound to be off quite a bit
        
         | liotier wrote:
         | InfoReseaux remarked that this data is suspicious, to say the
         | least: CC BY-NC 4.0 but contains ODbL licensed data coming from
         | Openstreetmap - but also from Microsoft.
         | 
         | Thread on the Openstreetmap forum:
         | https://community.openstreetmap.org/t/is-globalbuildingatlas...
         | 
         | Christian Quest sent the following message to the authors:
         | 
         | "I'm writing to you because I'm surprised by the choice of data
         | license you've set on the GlobalBuildingAtlas dataset.
         | 
         | As mentioned and explained in your paper, at least two data
         | sources you've been using to create this dataset are under the
         | Open Database License (ODbL): OpenStreetMap and Microsoft
         | building datasets.
         | 
         | I've downloaded the extract of data you're proposing to have a
         | look at the final dataset, and it confirms that building
         | polygons from OSM (and Microsoft) are present in the resulting
         | dataset in a substantial portion.
         | 
         | In such case, your dataset must be published under the ODbL
         | licence (see 4.2), because it is a derivative database (see 1.0
         | of ODbL license for definition).
         | 
         | A copy of this message has also been sent to the Legal Working
         | Group of the OSM Foundation.
         | 
         | Thanks in advance to fix quickly the license of the dataset you
         | published. This will also allow OpenStreetMap contributors to
         | use it to improve OpenStreetMap, which is not possible with the
         | CC-BY-NC you choose."
        
       | sl0wik wrote:
       | For sure not all buildings. My first query was "Costa del Este,
       | Juan Diaz, Distrito de Panama, Panama Province, Panama" and no
       | coverage. This region is packed with high rises.
        
         | whackernews wrote:
         | Where?
        
         | toss1 wrote:
         | Yup, check my home neighborhood and it is good coverage, but
         | checked a rural area I know has houses around a lake (not just
         | tiny fishing shacks, full-sized 200+m^2 houses) with addresses
         | and electrical and telecom/internet service, but there there
         | are no buildings shown. Plus, even though there is a lot of
         | tree coverage, many of the houses are clearly visible in Google
         | Satellite View, so it is unclear why they wouldn't have been
         | picked up.
         | 
         | Cool stuff, but they still have work to do.
        
           | fluoridation wrote:
           | The reason, I would guess, is that it takes at least a couple
           | satellite passes to be able to get the 3D data, and then a
           | processing step to actually obtain it, and they've not done
           | that for sparsely populated areas.
        
       | aswegs8 wrote:
       | Looks like it has trouble depicting the height of the highest
       | buildings like Burj Khalifa and Merdeka 118
        
       | bigfishrunning wrote:
       | All of the houses on my block are very weird shapes on this map,
       | it doesn't reflect reality very well
        
         | macintux wrote:
         | I haven't visited the map, but I wouldn't be surprised to learn
         | they didn't attempt to re-create the building shapes, just
         | location and estimated size. The real shape seems pointless for
         | this dataset.
        
           | wongarsu wrote:
           | Building shapes are actually pretty accurate in urban areas.
           | I haven't read the paper, but I think what they were after is
           | building volume. Which obviously requires the footprint shape
           | and height
        
         | oersted wrote:
         | If it was able to detect roughly how many buildings there are
         | in that area, that's already fairly useful at global scale. It
         | seems they are focused on measuring urbanization. It's good to
         | know though that it may be dodgy at that level of granularity.
        
         | freedomben wrote:
         | Indeed, my neighborhood doesn't show any buildings at all even
         | though I know all of our houses have been there at least 30
         | years and were still there when I left this morning.
        
           | dhosek wrote:
           | When you left this morning, sure, but who knows what's
           | happened in the interim.
        
       | plaguna wrote:
       | So here is finally the data that I needed for my idea: walk
       | directions from A to B while never leaving the shadows,
       | suggesting bars and pubs where to wait while the shadows catch up
       | and let you cross "safely" where before was sunny.
       | 
       | Useful for scorching weather places like south of Spain.
        
         | whackernews wrote:
         | I do a similar game when walking around but the complete
         | inverse. I live in the UK.
        
           | fluoridation wrote:
           | So running directions from B to A while never touching any
           | shade, suggesting churches and rehab centers to ignore while
           | the shadows stay perfectly still?
        
         | nceheil wrote:
         | You can fool me Dracula
        
         | maelito wrote:
         | Would need this, inverted, for winter.
        
           | szemy2 wrote:
           | Check out sunseekr its popular here in the uk:
           | https://sunseekr.com/
        
             | qwertox wrote:
             | In that case https://app.shadowmap.org could be a better
             | recommendation.
        
       | manoDev wrote:
       | It seemed quite accurate for my region (Sao Paulo downtown).
        
       | jabron wrote:
       | They've used a vision transformer to estimate building heights
       | from monocular aerial photographs, so they're guesses at best.
       | Calling this a map is a stretch.
        
         | chakintosh wrote:
         | I checked the building heights where I live vs the actual
         | heights, sometimes they display the height as half of what the
         | actual height is.
        
           | fainpul wrote:
           | I found a few that are off by a factor of about 4. Always
           | lower than in reality, never taller.
        
         | JoeAltmaier wrote:
         | Hey maps used to have sea serpents and winged griffins. It's a
         | map. Maybe not the best map.
        
       | snowstorm82 wrote:
       | Groups of boats in the dock have become buildings. Cool project
       | still.
        
         | suyash wrote:
         | lol, AI lacking effective QA process it seems
        
         | 1970-01-01 wrote:
         | If AI says a floating building is a building I will agree with
         | it. It isn't wrong but it isn't right.
        
           | wongarsu wrote:
           | I'd draw the line at boats that are primarily used when
           | docked. A house boat, a restaurant located on a permanently
           | docked boat or Russia's floating power plants are all
           | buildings in my book
        
       | mengmota wrote:
       | oh wow basically amazed
        
       | georgeecollins wrote:
       | I wonder when people will be able to make a map of all the people
       | in the world at a particular moment. Pretty soon I think.
        
       | devrundown wrote:
       | Would be great to somehow incorporate this into a game where you
       | can play in any city in the world.
        
       | dr_dshiv wrote:
       | Google recently has made efforts to link Gemini to Google Earth
       | data. "Earth AI"
       | 
       | https://ai.google/earth-ai/
       | 
       | Has anyone tried this? I'm not quite sure how to prompt it
       | effectively.
        
         | mistrial9 wrote:
         | arxiv:2510.18318
        
       | lo_zamoyski wrote:
       | The heights seem all over the place. Look at the WTC, for
       | example.
        
       | stevenjgarner wrote:
       | What is the crowd process to upgrade/correct this? I notice
       | massive errors in small Midwest US towns (multiple buildings
       | being combined/ommitted, etc)?
        
         | whizzter wrote:
         | It's machine-learning generated "slop" honestly.
         | 
         | Looking at where I live and where I grew up the building
         | heights are quite badly estimated.
         | 
         | - Some groups of houses around here that are more or less
         | identically built but on sloped terrain are reported to have
         | widly differing heights
         | 
         | - My neighbour building is reported to be half the height of
         | this building (they're more or less equally high at 5 stories)
         | 
         | - A small office shack behind the neighbour building is
         | reported to be taller than it (it's a single-story building,
         | the neighbour building is 5 stories)
         | 
         | - The freestanding buildings on the farm where I grew up are
         | like you said, badly combined, much of the estimation there
         | seems to be dependent on shadows,etc.
        
           | f4c39012 wrote:
           | Or, they subtracted a digital elevation model from a digital
           | surface model, ran a point-in-polygon match against an
           | existing building dataset, and labelled the difference as the
           | height of the building. No ML needed.
        
             | basscomm wrote:
             | There's a notice in the bottom-left corner on desktop that
             | says: "This is a machine-learning-derived product. Errors
             | may occur"
        
           | JoeAltmaier wrote:
           | Still useful for a sense of building density. If almost
           | everything called out is some sort of construction, then the
           | density map of the world is a realistic estimate of human
           | occupation.
        
         | AStrangeMorrow wrote:
         | Yeah some areas I looked at seemed fine (US, EU, Shangahi), but
         | some the buildings are on top of the roads, and they look more
         | like Perlin Noise than actual buildings. E.g main parts of
         | Beijing.
         | 
         | OpenStreetMap and such have pretty good building footprints,
         | sometimes enriched with building heights. Maybe that can be
         | used for some correction
        
         | cdkmoose wrote:
         | Looking around my home area, I found HS football bleachers
         | showing up as buildings and an outdoor basketball court showing
         | up as a building. Looking at the coloring of those features on
         | Google maps, I can see how the mistakes might have been made.
         | But still feels like it needs tuning.
         | 
         | To be fair, I checked my boyhood home in very rural Maine and
         | it was correct for the size and shape of the multiple farm
         | buildings.
        
       | dzink wrote:
       | The height measure when you click on a building is not very
       | accurate. Clicked on two next to each other (one single floor and
       | one with two floors) and the single story house was marked taller
       | by the site.
        
         | sdoering wrote:
         | Can confirm. My garage is marked double the height of my actual
         | house. Both are incorrect. My shop building on my property,
         | next to the garage is actually missing.
         | 
         | The positioning of the shapes in relation to each other are
         | also wrong. And not in a subtle way.
         | 
         | Looked at the house my father built in 1985. A few hundred
         | kilometers from my current location. The shape is wrong (as are
         | the shapes of all neighboring buildings. As are the positions
         | toward each other (distance between houses, rotation of
         | shapes). The heights are also significantly wrong. The two
         | story houses on the opposite side of the road are said to be
         | slightly above 2 meters in height.
        
         | d_silin wrote:
         | quite a lot of height errors, checked my neighborhood as well,
         | often by orders of magnitude off.
        
       | tppiotrowski wrote:
       | > This is a huge leap from the previous global dataset, which
       | contained about 1.7 billion buildings
       | 
       | I use the monthly release of Overture Maps building dataset and
       | the last one had around 2.3 billion buildings iirc
       | 
       | [1] https://docs.overturemaps.org/blog/2025/11/19/release-notes/
        
       | jspann wrote:
       | I've had a project idea in the back of my mind for a long time:
       | take all of my locations that I've saved in Swarm
       | (https://swarmapp.com) and 3D render all of the buildings that
       | I've been to. The issue was a lack of models of the actual
       | buildings - which is here! Maybe it is possible...
        
       | JoeAltmaier wrote:
       | Revealing: third-world areas look different. Curiously, Australia
       | resembles third-world building density more than it resembles the
       | industrialized world.
        
       ___________________________________________________________________
       (page generated 2025-12-10 23:01 UTC)