URI: 
        _______               __                   _______
       |   |   |.---.-..----.|  |--..-----..----. |    |  |.-----..--.--.--..-----.
       |       ||  _  ||  __||    < |  -__||   _| |       ||  -__||  |  |  ||__ --|
       |___|___||___._||____||__|__||_____||__|   |__|____||_____||________||_____|
                                                             on Gopher (inofficial)
  HTML Visit Hacker News on the Web
       
       
       COMMENT PAGE FOR:
  HTML   Skyfall-GS – Synthesizing Immersive 3D Urban Scenes from Satellite Imagery
       
       
        r0x0r007 wrote 11 hours 45 min ago:
        Seems like it could also be useful in planning combat(strategic)
        actions ad hoc, given limited resources, but I guess military already
        has some other tech for this...
       
        EagnaIonat wrote 15 hours 4 min ago:
        Pretty cool. Interesting that the shadows become 3 dimensional objects
        as well.
       
        nevster wrote 18 hours 14 min ago:
        I'm probably in a minority here. When I read Skyfall-GS, I immediately
        thought it was some new Apple IIGS game!
       
        CobrastanJorji wrote 23 hours 22 min ago:
        It looks good! I imagine a reasonable next step might be to do
        something about the cars, which are omnipresent in urban scenes but
        seem like they've been left a blurry mess in the examples.
       
        thicknavyrain wrote 1 day ago:
        This is so cool. I used to work on urban heat island analysis and now
        work in natural catastrophe modelling, and in both cases knowing the
        average heights/volumes of buildings is a very handy thing to have but
        is surprisingly difficult information to retrieve. Even a coarse
        estimate available at annual resolution has some really awesome use
        cases, very excited to see this.
       
        anigbrowl wrote 1 day ago:
        Is there any reason that this couldn't integrate Street View data?
       
          fsloth wrote 15 hours 45 min ago:
          For production scenarios data like that usually has a hefty license
          fee. Hence it’s interesting to find methods for scene generation
          from data that is cheaper to acquire.
       
        Qworg wrote 1 day ago:
        While the methodology wasn't published, Microsoft did something similar
        for Flight Simulator.
        
  HTML  [1]: https://www.smithsonianmag.com/air-space-magazine/flight-box-1...
       
          rkomorn wrote 1 day ago:
          If you want to dig more, I suspect Microsoft got a lot (if not all)
          of it from  blackshark.ai . They're one of the companies whose logo
          shows up during game start.
          
          I knew their name because, when I worked for an Airbus subsidiary, we
          talked with them about a solution to generate 3D environments for
          any/every airport.
          
          They had some cool stuff but also some wonky stuff at the time (like
          highway overpasses actually being rendered as walls across the
          highway).
       
            Stevvo wrote 22 hours 48 min ago:
            blackshark.ai was used for generating building models with textured
            facades from footprints. Their thing for generating airports was
            actually not used in MSFS. The photogrammetry from satellite photos
            in MSFS 2024 was done by Maxar Technologies.
       
              rkomorn wrote 16 hours 39 min ago:
              Ah, neat. How do you know this?
              
              And also, was it any different for MSFS 2020?
       
                Stevvo wrote 7 hours 41 min ago:
                My side-project is working on addons, so I spend a lot of time
                with Flight Simulator.
                2020 didn't have the photogrammetry from satellite photos, only
                from aerial photos in major cities. The blackshark buildings
                are used wherever they don't have any photogrammetry available.
       
                  rkomorn wrote 5 hours 9 min ago:
                  Neat (again :) ).
       
        aaroninsf wrote 1 day ago:
        Re: utility in games,
        
        I suspect hybrid solutions will remove the limitations of GS, with
        (eventually...) some smooth hand off. Do clean-enough GS like this;
        then hand the output to other systems which covert into forms more
        useful for your application and which adopt e.g. textures from
        localized photos etc.
        
        It's just a bit of engineering and compute...
       
        mtharrison wrote 1 day ago:
        Maybe dumb question but how do I just take a sat image and create the
        scene? The scripts in the repo are all about training which I assume
        requires you to have the 3d data too.
        
        These sort of projects always look cool but I think the real "wow
        factor" would be a file upload where you can see the result on your
        image. I assume there are reasons why this isn't done.
       
          pedalpete wrote 1 day ago:
          Do you mean converting your image into a 3D scene?
          
          This is where we were heading with our 3D volumetric video company
          [1] We were working on blending 3D satellite imagery with your ground
          view (or low flying in the case of paragliders) photos and videos to
          create a 3D scene.
          
          Our technology was acquired prior to us being able to fully realize
          the vision (and we moved on to another project).
          
  HTML    [1]: https://ayvri.com
       
        Mobius01 wrote 1 day ago:
        This would be the next step for flight simulators, which while
        remarkable still require handmade assets for accurate details.
       
          Stevvo wrote 1 day ago:
          MSFS 2024 already does photogrammetry from satellite photos. However,
          it builds triangle geometry much like is done from aerial
          photography, because gaussian splats are not suitable for games; you
          can't build collision geometry from a gaussian splat for example.
       
          zokier wrote 1 day ago:
          afaik msfs uses partially automatically generated 3d assets (from
          Bing Maps?) from aerial/satellite imagery.
       
        wkat4242 wrote 1 day ago:
        Nice, but when you look up close things like this and Google Earth look
        like a post-apocalyptic scene :)
        
        It would be amazing if they could also take user-generated photos and
        videos at ground level and accurate mapping data (that has building
        outlines) and clean that up to something presentable.
        
        I mean, what they do here is what google and apple are already doing
        for years. It's time for the next step.
       
          zokier wrote 1 day ago:
          > I mean, what they do here is what google and apple are already
          doing for years
          
          This is gaussian splatting. I'm pretty confident that google/apple
          have not done that.
       
            wkat4242 wrote 1 day ago:
            Oh I didn't realise it was a different technique but the result is
            similarly bad when zoomed in :(
       
        marcodiego wrote 1 day ago:
        This could be specially good for a world 3d model for flightgear.
       
          anthk wrote 1 day ago:
          Once Flightgear could use Google Earth assets.
       
        voidUpdate wrote 1 day ago:
        "explorable" and "immersive" is definitely a bold choice of words when
        you can't really get below the level of the buildings before the
        gaussian splatting is very obvious. Sure, it's impressive that you can
        get that detailed from a few satellite images, but I think that might
        be overselling it a bit
       
          Animats wrote 18 hours 40 min ago:
          > you can't really get below the level of the buildings before the
          gaussian splatting is very obvious.
          
          True. If you bring the viewpoint down to near street level and look
          horizontally, it's worse than traditional photogrammetry methods.
          
          I've been looking for algorithms like this for representing distant
          regions in virtual worlds. Open Drone Map can do a good  job,
          sometimes, but it really needs a cleanup pass.
       
          makeitdouble wrote 23 hours 46 min ago:
          That's a matter of data, and this looks promising to me. If instead
          of satellite images they'd feed it drone shots they could probably
          get down to a level of detail that becomes actually immersive and
          would be way beyond the digital twins we currently have.
       
          echelon wrote 1 day ago:
          We're early days. Models will soon interpolate all of that.
          Eventually in real time.
          
          I wouldn't knock the research. The results look impressive to me.
       
            Skyy93 wrote 1 day ago:
            We probably won't. GS is a reconstructive method, so when data is
            unavailable, you can only perform poor interpolation. You would
            need additional generative, not reconstructive, models. However,
            this would open the door to unfaithful augmentation again.
       
              echelon wrote 1 day ago:
              Different applications.
              
              GIS won't want generative hallucinations.
              
              Consumer mapping apps, social applications, and games (eg. flight
              sims) will want the maps to look as good as possible.
       
                wkat4242 wrote 1 day ago:
                GIS don't want half exploded buildings either. Nor would they
                care about photographic textures on the 3D models.
       
                  fsloth wrote 1 day ago:
                  You’d be surprised what GIS - or at least GIS - adjacent
                  customers want. If you think about any cute-but-useless map
                  detail that comes to your mind there is likely a paying
                  customer for it.
       
        daemonologist wrote 1 day ago:
        Very cool; interesting how it turns all the trees into puffballs
        though.  Some artifact of the pre-trained depth estimation or diffusion
        model maybe?
       
        p0w3n3d wrote 1 day ago:
        Now the GTA: Anywhere please...
       
          Y_Y wrote 1 day ago:
          Ya, can't wait to play GTA: Nova Zemyla.
          
          In fact you wouldn't even need to be limited to earth. Why not throw
          in Google Moon and steal a moon buggy while shooting scientific
          rovers and doing cool flips out of craters?
       
            iammjm wrote 1 day ago:
            There are already 3D globes of the moon, for example with cesiumjs:
            [1] you can even import it to unity and other engines
            
  HTML      [1]: https://sandcastle.cesium.com/?id=moon
       
       
   DIR <- back to front page