Live data from Hacker News

Google Photorealistic 3D Tiles and Unreal Engine

nilsbakker.nl

71–80 of 131 posts

Re: Google Photorealistic 3D Tiles and Unreal Engine

#71
post #8

HOW DO I PLAY?? (serious)

Microsoft Flight Simulator. (serious)

It's a bit the same story as GPT:

Early: Google does primitive tech (Google Earth)

Later: Microsoft does amazing implementation (Flight Simulator / Bing)

Finally: Google tries to copy but fails (Google Maps)

Re: Google Photorealistic 3D Tiles and Unreal Engine

#72
post #39
post #35

For the OP: Warning: Potential Security Risk Ahead Firefox detected a potential security threat and did not continue to nilsbakker.nl. If you visit this site, attackers could try to steal information like your passwords, emails, or credit card details. And: $ curl https://nilsbakker.nl/portfolio/3d-tiles/ curl: (60) SSL certificate problem: self-signed certificate More details here: https://curl.se/docs/sslcerts.html…

I'm getting an error as well, although not a self-signed error. I'm getting a mismatched domain error.

Probably just depends what order your TLS implementation is deciding to complain about the issues: it's not only self-signed but has no DNS SAN and the CN also mismatches.

Re: Google Photorealistic 3D Tiles and Unreal Engine

#77

For years I've been dreaming of a driving/pedestrian sandbox game that used Google Maps data to let me drive recklessly around my home town with accurate models of the buildings and roads. This demo reminds me of that dream.

Me too. The problem seems to be mainly that the models are still awfully approximate and the textures are horribly blurry. They're fantastic for a bird's-eye view or flight simulators, but not what you want for driving.

I'm hoping that at some point in the next 10 years, Google/Microsoft come up with a high-res version produced mainly by Street View-type data or similar drone-supplied data. I suspect that basic photogrammetry is the easy part -- that the challenging part is dealing with changing sun position, hourly/daily/monthly weather/seasons, and all of the pedestrians and moving/parked vehicles.

Re: Google Photorealistic 3D Tiles and Unreal Engine

#78
post #33

Earlier quoted context omitted.

To this day this is the best experience I had in VR by far. I spent hours roaming the earth, it's truly amazing. It's sad that Google is massively underutilizing this asset.

What should Google be doing with it?

Team up with microsoft and asobo.

Re: Google Photorealistic 3D Tiles and Unreal Engine

#79
post #38

Earlier quoted context omitted.

These are the fields on the certificate that's being presented to me: emailAddress = root@dns-redirect001.axc.nl, CN = dns-redirect001.axc.nl, OU = SomeOrganizationalUnit, O = SomeOrganization, L = SomeCity, ST = SomeState, C = -- A self-signed certificate that expired on 6/11/19, 4:35:46 PM GMT+2 It's probably been taken down.

It works for me (again) but I remember that server domain from my old Versio shared hosting account. They used to be the typical "dirt cheap, low spec, PHP 5.4" hoster of choice for students and people looking to save a buck, but they've since raised their prices significantly (and I don't think they've raised their quality of service much). Would not recommend. I wouldn't recommend shared hosting in the first place,…

Github pages is excellent and should be the dev’s go to of choice IMO.

Re: Google Photorealistic 3D Tiles and Unreal Engine

#80

For years I've been dreaming of a driving/pedestrian sandbox game that used Google Maps data to let me drive recklessly around my home town with accurate models of the buildings and roads. This demo reminds me of that dream.

Me too. The problem seems to be mainly that the models are still awfully approximate and the textures are horribly blurry. They're fantastic for a bird's-eye view or flight simulators, but not what you want for driving. I'm hoping that at some point in the next 10 years, Google/Microsoft come up with a high-res version produced mainly by Street View-type data or similar drone-supplied data. I suspect that basic photo…

AI upscaling with diffusion models will probably solve that problem.
Post reply on HN