Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!iuvax!rutgers!orstcs!ogccse!blake!uw-beaver!ubc-cs!alberta!calgary!ctycal!ingoldsb
From: ingoldsb@ctycal.COM (Terry Ingoldsby)
Newsgroups: comp.graphics
Subject: Re: 3d Computer Generated Holography
Summary: 2D or 3D Fourier?
Keywords: What's happening in the field?
Message-ID: <448@ctycal.UUCP>
Date: 25 Aug 89 18:31:45 GMT
References: <441@ctycal.UUCP> <1306@blackbird.afit.af.mil>
Organization: The City of Calgary, Ab
Lines: 50

In article <1306@blackbird.afit.af.mil>, news@blackbird.afit.af.mil (News System Account) writes:
> The research at MIT, that I`m aware of, is being conducted by Steve Benton.
> He's producing Synthetic Holographic Stereograms.  In general they produce
> a series of perspective views using convential computer graphic techniques.
> Each image is then projected with laser light onto a sheet of holographic
> film from the angle corresponding to its computed viewpoint.  The effect
> is to provide the visual properties of an actual hologram.  This works because
> each eye receives a different perspective view.  As you move around you get
> the stereoscopic effect, hence 3-D.
In other words, the same as `Leslie' hologram referred to by another poster.
Also similar to a white light space shuttle hologram I saw a few years ago.
> 
> There have been several articles written in the SPIE journals.  Volume 
> 761 has one titled "Alcove Holograms for Computer_Aided Design".
> 
> As far as work being done on actual 3-D CGH, the Air Force Institue of
> Techology is doing some work in this area.  They are using points to 
> describe an object and then calculating the light propogated from
> a point on the object to every point in the hologram plane.  This
> is being done without the use of Fourier analysis.  Anyway, the 
> calculated intensity pattern is plotted out and photo-reduced
> onto a holographic plate.  
Do you know if they use 2D or 3D Fourier?  When I made my 2D holograms
I basically used a 2D FFT to calculate the diffraction pattern of an
aperture, mixed in a reference beam and output the resulting pattern
to a monitor.  There I photographed it (and reduced its size) so that
I could reproduce it using a laser.  Note that the laser is not
necessary for reproduction; it can be reproduced in the computer but
its not as much fun.

A professor I had once mentioned that he thought that 3D holography
worked because film emulsion was 3D (in comparision to the wavelength
of light).  I'm not sure if he was correct.  If that is true, do I have
to use a 3D FFT?  How do I get it on the emulsion?  Since a normal
hologram does not give a full 3D (ie. 360 degree) image could the
calculations be restricted (ie. not use a 3D FFT).

I can also see how a sort of super glorified ray-tracing effort
would allow you to calculate the interference of all the light
reflected from a 3D object, but surely the calculations would
require more computer power than is currently available.  (This
is as much a question as a statement).




-- 
  Terry Ingoldsby                       ctycal!ingoldsb@calgary.UUCP
  Land Information Systems                           or
  The City of Calgary         ...{alberta,ubc-cs,utai}!calgary!ctycal!ingoldsb