Transform a 2D color image into 2 images of (R1,G1,B) at each pixelof image 1 and (R2,G2,B) at eac

[88888 Dihedral]

I am wondering is there any open source C++ lib can do this.

Could any function in OpenGl do this non-trivial task?

Or I have to do it from the groud up again?

 

[Juha Nieminen]

I am wondering is there any open source C++ lib can do this.

Could any function in OpenGl do this non-trivial task?

Or I have to do it from the groud up again?

Please do not ask your question in the subject line, especially given that
it's such a long and complicated one.

 

[Victor Bazarov]

Please do not ask your question in the subject line, especially given that
it's such a long and complicated one.

....and may I add a suggestion to not ask the question in the subject
line in the right newsgroup, as well? comp.graphics.api.opengl is the
right one if you need OpenGL, or just comp.graphics.algorithms if you
need somebody's help on devising one (algorithm, that is)

Thanks!

V

 

[88888 Dihedral]

...and may I add a suggestion to not ask the question in the subject
line in the right newsgroup, as well? comp.graphics.api.opengl is the
right one if you need OpenGL, or just comp.graphics.algorithms if you
need somebody's help on devising one (algorithm, that is)

Thanks!

V

Thus, there is no easy open source for this problem in C++.
Am I guessing wrong ?

 

[Victor Bazarov]

Thus, there is no easy open source for this problem in C++.
Am I guessing wrong ?

<shrug> I have no idea. But if *I* were to try to find a solution, I'd
be looking *elsewhere*. comp.lang.c++ is not a repository of all open
source solutions for all possible problems that can be implemented in C++.

V

 

[88888 Dihedral]

<shrug> I have no idea. But if *I* were to try to find a solution, I'd
be looking *elsewhere*. comp.lang.c++ is not a repository of all open
source solutions for all possible problems that can be implemented in C++.

V

If a language does not have a smart library then it is really tedious
to do everything from the ground up again.

OK, I'll use the Simple-Directmedia Library to play with real 3D stereo
graphics.

 

[Jeff Flinn]

If a language does not have a smart library then it is really tedious
to do everything from the ground up again.

OK, I'll use the Simple-Directmedia Library to play with real 3D stereo
graphics.

See boost's Generic Image Library at
http://www.boost.org/doc/libs/1_48_0/libs/gil/doc/index.html.

Jeff

Jeff

 

[LR]

See boost's Generic Image Library at
http://www.boost.org/doc/libs/1_48_0/libs/gil/doc/index.html.

Jeff

You may also find http://opencv.willowgarage.com/wiki/ useful.

 

[Victor Bazarov]

If a language does not have a smart library then it is really tedious
to do everything from the ground up again.

Generally speaking, there is no such thing as "a smart library" for
every possible future problem that people haven't yet encountered, not
to mention to satisfy every possible requirement from every potential
user of such a library, so there *always* be somebody contesting the
"smartness" of that library.

OK, I'll use the Simple-Directmedia Library to play with real 3D stereo
graphics.

Best of luck with it!

V

 

[88888 Dihedral]

Generally speaking, there is no such thing as "a smart library" for
every possible future problem that people haven't yet encountered, not
to mention to satisfy every possible requirement from every potential
user of such a library, so there *always* be somebody contesting the
"smartness" of that library.


Best of luck with it!

V

I have to separate the background still objects and the moving or interactive foreground objects such as humans and animals first.

Thus a good and fast edge tracing algorithm is necessary.

Any suggestions?

I have to perform edge tracing of sub-sampled images first.

The full resolution images will give more details.

 

[Victor Bazarov]

I have to separate the background still objects and the moving or interactive foreground objects such as humans and animals first.

Thus a good and fast edge tracing algorithm is necessary.

Any suggestions?

I have to perform edge tracing of sub-sampled images first.

The full resolution images will give more details.

I don't have any suggestion on a tracing algorithm or library. I have a
suggestion, though. Since you're interested in graphics and image
processing, post to 'comp.graphics.algorithms', I am sure you will find
more information that way.

V