A Map of Every Building in America — The New York Times


#1

From The New York Times:


#2

I created a simple plugin to read these files. The file for Washington D.C. was the smallest, so it was my test case. Using Google Earth, I obtained the GPS coordinates to bracket the area I wanted:

The placemarks give me the exact locations (I converted these to signed decimal notation):

X minimum (Longitude): -77.034522
Y minimum (Latitude): 38.886533
X maximum (Longitude): -77.015675
Y maximum (Latitude): 38.897192

This is the result:

The plugin can be found at: https://sites.google.com/site/spirixcode/code/jimhami42_geojson_importer.rbz


#3

Trying to get it to work with a section of the Wisconsin json. No luck after multiple attempts.

Tried the coordinates shown below.

Get:
Screenshot%20-%2010_13_2018%20%2C%2012_08_02%20PM

There are clearly buildings shown on the map, though.
Screenshot%20-%2010_13_2018%20%2C%2012_10_46%20PM


#4

Try switching your max/min Longitudes (-88.220932 is less than -88.216927).

This is definitely a memory hog. I imported part of Milwaukee … it took a l-o-n-g time with only 8GB of memory (SketchUp wanted 10GB+).


#5

I just realized the difference between your longitude entries and mine. So the coordinate order should be SW corner and NE corner?

It does take a while to get the Wisconsin json opened.


#6

Yes … I didn’t make that clear at all ;(


#7

It was clear once I really looked at the values in your entries. Maybe you could change the labels to be SW long, SW lat, NE long, NE lat.

By the way. It works great when you enter the numbers correctly. :smiley:


#8

May I make a feature request? Could it also geolocate the model so the resulting KMZ would display in the correct location in GE? That would just require writing the SW coordinates to the Geolocation.


#9

I added that to the plugin (you can download it at the same location).

It actually lines up fairly well:


#10

I’ll give it a try. I manually entered the SW coordinates for geolocation and found it lined up pretty accurately, too.


#11

That’s what I like about the integration of Ruby into SketchUp and being able to add your own little plugins on the fly (an open system). For some people/cases this plugin might be a big time saver.
edit: spelling


#12

The building footprints are created from satellite images using a variety of algorithms to deduce the polygonal shape:

image

One of the buildings where I work looks like this:

image

This is what is in the geoJSON file:

image

Not especially comforting :neutral_face:

There is a field in the file for “properties” that is currently unused. It would be nice if this is eventually populated with building heights (or other data).


#13

NYC 3-D Building Model
https://www1.nyc.gov/site/doitt/initiatives/3d-building.page


#14

A couple of years ago, I imported NYC into SketchUp. At the time, the data-set didn’t provide any roofing/facade detail … this new data-set offers a lot more potential detail :slight_smile: