🚀 go-pugleaf

wright@lims01.lerc.nasa.gov (Ted Wright) writes:
>willmore@iastate.edu (David Willmore) writes:
>>There are a few differences.  The wavelength of the radio waves in
>>question are much longer than the light waves.  Therefore, the precision
>>with which the image receivers must be places is more stringent for the
>>optical methods.  It's just not too easy.  I'm not saying that it 
>>can't or isn't being done.  It's just difficult to accomplish.

>A few years ago on a television program (NOVA, I think) I saw something
>like this. Some people at the Jet Propulion Lab took a couple of digitized
>satellite pictures, crunched them in a computer for a while, and ended up
>with a 3D model of the subject area (San Francisco, I think) in their
>computer.

>They could "fly" around the city (between building, etc) and view
>objects from different perspectives. I don't think the were increasing
>the resolution, but they were extracting the 3rd dimension information
>from 2D images.  

I work at a GIS (geographic information systems) laboratory where we
work with data like this.  There is a comercial package called pv wave
which does this.  One of their demos takes several satellite pictures
from different angles and combines them to get a topography for the
area.  Given that, they map a set of luma (greyscales) values to the
topo yielding a 3D model of the city.  It's impressive to see a two year
old workstation rotate a 3D model of LA in real time. :)

I'm afraid that we've wondered afar, sorry.

--David Willmore
willmore@iastate.edu