CyberCity 3D and OpenStreetMap: A Comparison

In the 3D modeling world, there are very few software tools available for large-scale city models. It is HARD to build an accurate model of every single building over a wide area, and such efforts require a significant investment of time and money. CyberCity 3D solves this through patented technology and an efficient production methodology to create models relatively quickly and affordably. Of course, there are some alternative solutions, one of which employs a very powerful resource — the public. This resource, however, does have its drawbacks.

OpenStreetMap is an open-source — and more importantly, crowd-sourced — map of the world. Over 2 million users have collected, created, and curated data to be made available over the Open Database License. Inspired by other crowd-sourced databases like Wikipedia, users can register on the OpenStreetMap website and start editing the map within minutes. In addition to their popular 2D map, OpenStreetMap also offers crowd-sourced 3D building data that can be used on 3D platforms including two that we work closely with — Cesium and InfraWorks 360.

OpenStreetMap buildings in Miami (as seen on opensciencemap.org)

OpenStreetMap has several beneficial uses, not the least of which is offering greater access to spatial information in regions of the world that otherwise do not have accurate map data. There are, however, some important flaws when using this data within the high-resolution demands of the AEC and GIS worlds. These differences are important when given the choice between crowd-sourced building information and CyberCity 3D photogrammetric models. Let’s take a brief look at this disparity:

Data Quality: As crowd-sourced data, the quality of building geometry and information varies greatly for OSM buildings, depending on the source. Most of the models are generated as basic extrusions of building footprints raised to assigned height values, or simply boxes drawn with the naked eye. Both methods offer very little real-world accuracy, and hold essentially no value for street-level visualizations. Bird’s eye view “flyovers” don’t require such high accuracy geometry, but the moment map users need to zoom into a specific street corner or block, these buildings can potentially confuse and mislead users about the realities of the visualized environment. All CyberCity 3D models, on the other hand, are created using photogrammetry of stereo-imagery. This ensures that the dimensions of our buildings are very accurate — usually within 6 inches. From high altitude flyovers to street-level analysis, CyberCity 3D models reflect real-world architecture at a level of detail not seen with OSM.

CyberCity 3D buildings in London compared to OSM models (as seen in Autodesk InfraWorks 360)

Data Coverage: OSM coverage is extensive in large city centers. With greater interest and availability of public data in these areas, models are usually more detailed here. Outside of metropolises, data coverage is much more limited. CyberCity 3D currently has a 3D library of over 90 cities across the globe. The key difference is that with the ability to model from imagery captured by satellite, CyberCity 3D can model any area in the world, on demand, with consistent quality.

CyberCity 3D Library — Growing by the Day

Data Export & Licensing: There are some options for exporting OSM building data, depending on the site hosting the data, but this is an area where CyberCity 3D shines. Every model in our library, no matter the order size, is available in the most popular 3D formats. Once ordered, our data is typically offered for unlimited use within an organization, too. Whether you need to work with data in InfraWorks 360, Cesium, SketchUp, ArcGIS, or Google Earth, we have you covered!

CyberCity 3D buildings are optimized for popular 3D modeling applications

Hopefully this post can bring some clarity to your decisions around 3D efforts. There are other differences, of course, but we focused in on the major points in our comparison. This is also not to say that one is always better than the other — we work with clients that aim to combine the 2 data-sets to leverage the strengths of both. As always, contact us with any questions you might have. Happy mapping!

Written by Austin Logie