 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I know it's a long shot but...
Does anyone know of a way to take a normal JPG, possibly scanned in
using a 30-bit scanner, and "reverse engineer" it into a set of objects
readable by POV or Moray? I just ordered Photoshop and Illustrator..
would Photoshop do something like this?
-mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Nope.
This is along the lines of those "Computer Enhancements" shown on TV shows
that magically make someone's face appear out of a muddy, blurred photo...
: )
-aardvarko
Ken Penner wrote in message <35562E2D.8AA221F7@worldnet.att.net>...
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Just a wild notion that occuredto me, but maybe you could use something
along the lines of OCR to do it. I've never heard of a program that would
do this, but if you could make one that was capable of analyzing the shapes
in a pic. (like the find edges function in paint programs) maybe you
could.
But, to answer your question, I'm pretty sure that right now there are
no such programs on the market..
KillFile
Ken Penner wrote:
> I know it's a long shot but...
>
> Does anyone know of a way to take a normal JPG, possibly scanned in
> using a 30-bit scanner, and "reverse engineer" it into a set of objects
> readable by POV or Moray? I just ordered Photoshop and Illustrator..
> would Photoshop do something like this?
>
> -mike
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ken Penner <kpe### [at] worldnet attnet> wrote:
: Does anyone know of a way to take a normal JPG, possibly scanned in
: using a 30-bit scanner, and "reverse engineer" it into a set of objects
: readable by POV or Moray? I just ordered Photoshop and Illustrator..
: would Photoshop do something like this?
That's completely impossible.
When you project from 3D to 2D, you loose information.
- There is information in the 3D space which ends up hidden in the 2D
projection, and you have no way to get that information back, apart
from guessing (artifficial intelligence?) (and even human guessing is
far from perfect).
- The projection is not unambiguous: Completely different 3D-objects may
produce identical 2D-projections. You have no way to tell which one of
those objects was the original 3D-object.
- The projection is not unambiguous even if you have several projections
of the same 3D-object (at least not always). Of course several projections
make guessing easier.
- You don't even have depth information. There may be two pixels in the image
with identical colors, but one of them may be extremely near to the camera
and the other at 1 billion kilometers. You have no way to know that (apart
from guessing).
--
- Warp. -
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <6kuglv$41u$3@oz.aussie.org>,
Nieminen Mika <war### [at] assaricctutfi> writes:
> That's completely impossible.
Eh, I wouldn't quite agree with you on that... under cirtain circumstances
of course.
From just a single JPG, no. That is for the guys in Mission Impossible to
play with. 3D requires *LOTS* more information then 2D. In theory,
infintally more. In reality, it depends on your resolution. If you
have a 1000x1000 picture and want about the same resolution in 3D, you
need 1000 times as much data. You simply can't magically extract that much
info from a single picture unless there are very special circumstances.
In the movies however, anything is possible.
From two photos however you might do it. The computer can, just like the
human brain, extract depth information if it has two pics to work with.
I accorance what I said in the previous paragraph, this method will not
give you 1000 times as much info, it will just detect the surfaces for you.
/Michael
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
