Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D

Notice
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
<-- Date Index --> <-- Thread Index --> [Author Index]

P3D Re: single shot-single lens 3d digital camera


  • From: Dan Vint <dvint@xxxxxxxx>
  • Subject: P3D Re: single shot-single lens 3d digital camera
  • Date: Thu, 27 Jan 2000 08:50:59 -0700

I think this was the same camera and software combination that I recently
saw but it was only $5000.

What is the relation to Stereo - simple, once you have a 3d model in the
computer you can now animate and use multiple cameras to generate stereo
images - this is what Boris does when he isn't working with his nodes and
real cameras.

In the little description I saw on this camera it was any object could be
photographed and converted. The grid like you say allows the software to
calculate the points in space and if you do a sequence of images/points
consistently you can get the full model. The choice of faces I think comes
from trying to build realistic faces with geometric shapes unless you have
specific software for the task this isn't all that easy to do with CAD like
software. But the system shouldn't be limited to this unless that was their
only intended market.

..dan

At 08:41 AM 1/27/00 -0700, you wrote:
>I'm coming back to this thread (2d->3d conversion) because once ago I surfed
>on a web site advertising about a device (digital camera+ flash) allowing to
>capture a human face in 3d, just in a single shot.
>The related software was claimed able to automatically generate the digital
>3d "portrait" in a few minutes.
>I asked myself  some questions: why flashlight should project a grid? why
>faces? I didn't find an answer and nealy forgot the problem, until I found
>on this list some postings about 2d->3d conversion. Now I think I've got the
>answers.
>Why the grid? probably because software reads the parallax of each of the
>grid's nodes from the 2d image:if camera's and flash positions are known, i
>shouldn't be difficult to (grossly) rebuild in 3d  the photographed surface.
>Why faces? probably because the rendering a human face strikes the
>attention, but  fractal models of human faces are less difficult to build
>than they could seem at first glance: just a 3d police identi-kit! Even the
>range of colours  is quite limited, so final rendering could give the
>impression of nearly-photo quality. Not enough for
>$10,000!
>I wonder if I guessed, but probably I did. Anyway  this is VR, not
>stereoscopy.
>Sergio
>
>


---------------------------------------------------------------------------
Danny Vint                                       http://www.dvint.com
Author: SGML at Work   	http://www.slip.net/~dvint/pubs/sgmlatwork.shtml 
                                                       mailto:dvint@xxxxxxxx
    
Calian                                              Voice:804-975-3482
mailto:d.vint@xxxxxxxxxx
Charlottesville, VA Office                http://www.calian.com