Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

Opus 5.5 helped me build Tokyo in the browser with Three.js by compiling open data - PLATEAU's 3D city model, OpenStreetMap and GSI elevation and aerial photos into streaming tiles, with procedural shader facades, generated road markings, signal-obeying traffic and trains. Also added Takram's volumetric clouds which cast shadows...

27,849 görüntüleme • 11 gün önce •via X (Twitter)

17 Yorum

jeantimex profil fotoğrafı
jeantimex11 gün önce

Source code:

Raghav Katta profil fotoğrafı
Raghav Katta11 gün önce

@threejs how are the different source datasets aligned at tile boundaries? a route crossing a seam seems like a useful test for elevation offsets, road continuity, and traffic handoff, especially when adjacent tiles arrive at different times.

jeantimex profil fotoğrafı
jeantimex11 gün önce

@threejs Good test idea! Tile seams turned out fine by design: elevation is one global 5 m grid and the road graph is global, so there's no handoff between tiles. But your question led me to a real handoff bug: cars leaving a flyover dropped to the ground under the ramp. Fixed now 🙏

1Broom profil fotoğrafı
1Broom11 gün önce

@threejs Cloud shade sliding across real neighborhoods sells the whole thing. Pulling public mapping sets into something this smooth is a serious amount of plumbing, and it shows.

mmsdaus profil fotoğrafı
mmsdaus11 gün önce

@threejs love it! i have this on my list to try with blender then export to unity

jeantimex profil fotoğrafı
jeantimex11 gün önce

@threejs looking forward to it!

ThisMightWork profil fotoğrafı
ThisMightWork11 gün önce

@threejs This makes me want an 'under the hood' toggle: buildings, terrain, roads, then the finished city. Seeing what each public dataset contributes could be almost as interesting as exploring Tokyo.

Aeravela profil fotoğrafı
Aeravela11 gün önce

@threejs looks good

jeantimex profil fotoğrafı
jeantimex11 gün önce

@threejs But it might be slow, I only test it on 5090 :)))))) Anyway, we can try to improve it

LA8 profil fotoğrafı
LA811 gün önce

How did the PLATEAU buildings line up with the GSI terrain? Curious whether the heights matched directly or needed corrections around the building bases.

jeantimex profil fotoğrafı
jeantimex11 gün önce

They matched directly: both PLATEAU and the GSI elevation tiles give heights above Tokyo Peil (the Japanese vertical datum), so no datum correction was needed, only small fixes at the bases. Shibuya is worse because it is hilly: a building on a slope has one base height but the ground under it varies by a meter or two across the footprint. Base height comes from PLATEAU, not the terrain. Each building sits at the lowest point of its own LOD1 solid. Only buildings with no solid fall back to the terrain height.

Tsuyoshi🛠️Swift-Slide|AI個人開発 profil fotoğrafı
Tsuyoshi🛠️Swift-Slide|AI個人開発11 gün önce

@threejs これはすごい。 試してみましたが本当に構築できてました。 これを活用していろんなことができそうです。

Lee Wyatt Corp profil fotoğrafı
Lee Wyatt Corp11 gün önce

@threejs Signal-obeying traffic is the detail that sells it. Everything else is just buildings.

dazacode profil fotoğrafı
dazacode11 gün önce

@threejs how much tokens did you used to make this possible?

jeantimex profil fotoğrafı
jeantimex11 gün önce

@threejs 3 hours of work, haven’t paid too much attention on token usages :)

INGENIUMOWL profil fotoğrafı
INGENIUMOWL11 gün önce

@threejs I like it. Open data into a living Tokyo.

Ayhan KORKMAZ profil fotoğrafı
Ayhan KORKMAZ11 gün önce

@threejs Very good !

Benzer Videolar

Hi Friends, for my first project created with GPT-6 Astra, I built the entire city of Seoul in 3D—a small revolution in digital geography! 😄 This interactive 3D miniature covers all of Seoul and its surrounding areas, and you can explore it yourself. It includes: • All 25 districts of Seoul and approximately 267,000 simplified building models • Rotate, zoom, and pan controls, plus an automated fly-through of 14 landmarks • Day, sunset, and night modes • Mobile touch support The buildings were generated from real-world map data, with their shapes simplified for performance. Terrain and building heights are exaggerated by 4× to make the cityscape easier to see. The entire process—from creating the project to completing deployment—took 43 minutes and 38 seconds minutes. Please note that the buildings are simplified models based on locations and footprints recorded in OpenStreetMap. Where height data is unavailable, estimated heights derived from the source data are used. The landmark models are interpretive recreations designed to emphasize their distinctive features, rather than survey-grade or photorealistic reconstructions of every building. The project is based on real road, waterway, elevation, and building data. • Maps and buildings: OpenFreeMap / OpenMapTiles / © OpenStreetMap contributors · ODbL Data snapshot: August 30, 2026 • Terrain: AWS Terrain Tiles / Mapzen · SRTM and other sources • District boundaries: southkorea/seoul-maps · Statistics Korea’s 2013 boundaries These may differ from Seoul’s current administrative boundaries. • 3D rendering: Three.js • Additional components: terrain, road, and building data, subject to their respective licenses

synabreu

338,027 görüntüleme • 1 ay önce