Pedro Martins emailed us a link to the ‘CityEngine’ by Pascal Mueller a PhD candidate and research assistant at the Computer Vision Lab of the ETH Zurich, Switzerland.
The CityEngine is focused on the procedural modeling of buildings and brings together the term ‘Automatic Architecture’ that is becoming increasingly common in the industry. We will have more on Automatic Architecture and work coming out research labs in Japan in a future post.
The YouTube demo reel below details Pascal’s work on Pompeii and general cityscapes, its it both fascinating and impressive at the same time – bearing in mind all of this is automatic:
Take a look at the CityEngine wiki for more info, including SIGGRAPH papers.
Its not surprising to hear that Google is interested in procedural city modelling techniques and i would expect to see Pompeii and other historic reconstructions in Google Earth sometime soon.
Time has got the better of us this evening but the test import of the Virtual London model into Crysis is going well – the screen grab below illustrates the London Eye imported from 3D Max in Crysis:
Time has got the better of us this evening but the test import of the Virtual London model into Crysis is going well – the screen grab below illustrates the London Eye imported from 3D Max in Crysis:
Duncan Smith, an up and coming PhD student at CASA, has been putting Land eXplorer through its paces these last few weeks. In the spirit of the centre he kindly produced a blog post and a movie of his thoughts:
The aim of digital cities is to combine photorealistic urban visualisations with rich analysis tools for built environment professionals and the public. We’ve seen some exciting advances towards this goal recently, but we’re not quite there yet. Right now it’s difficult to make the link between the analysis and database functions of GIS software, and the photorealistic visualisations that 3D graphics software provides. 3D GIS applications like ESRI ArcScene are a step in the right direction, but don’t yet offer the 3D graphics power, or web publishing functions, that ideally we’d like to see.
The software developer 3D Geo GmbH, a spin-off of the Hasso Platter-Institute for system technique (HPI), has recently released their LandXplorer software to allow the creation of 3D city models that combine GIS functions with photorealistic graphics. LandXplorer was used to create the excellent Berlin and Dresden 3d models published in Google Earth. 3D Geo are also a key developer in the CityGML standard for 3D urban data. We’ve been testing LandXplorer Studio and are impressed with its features.
The YouTube movie embeded below shows some highlights:
Very large terrain datasets can be viewed and explored in 3D. We managed to load in all of Greater London at 4m resolution, and our Stratford data at 1m resolution. It is straightforward to import shapefiles of building outlines to create basic 3d city models on top of this terrain layer. The building models can be classified with attributes as the Stratford example shows. You can skip between the 3D attribute view and phototextured views, with the software handling aerial photography well and allowing the importing of textured buildings in 3ds format.
Overall LandXplorer is an innovative piece of software that brings functional digital cities a step closer. Currently it doesn’t offer much in the way of spatial analysis, but 3D Geo as ESRI have recently become integration partners so this should improve in future releases.
Take a look at the Land eXplorer home page for a free viewer and a ‘try before you buy’ demo.
Pedro Martins emailed us a link to the ‘CityEngine’ by Pascal Mueller a PhD candidate and research assistant at the Computer Vision Lab of the ETH Zurich, Switzerland.
