[Fwd: panoramic pictures color adjustments]

Hi all,

Phillipe Thomin as send me the following email, I'm forwarding it because it
might be useful for the PTmender development as well as for further
additions of the nona stitcher.

From: Philippe Thomin <philippe.thomin <at> univ-valenciennes.fr>
Subject: panoramic pictures color adjustments
Date: 2006-01-04 10:58:40 GMT

Dear Pablo,

I use hugin for some month, and I find it very comfortable compared to 
hand-made script editing for Dersch's Panorama Tools.

I am using an amd64 computer,so I do not longer use the old 32 bits 
Panorama Tools library, so I loss the color correction capabilities it 
offered.

This is why I wrote a small utility to do color adjustements. I followed 
the ideas exposed by H. Dersch in his article at 
www.path.unimelb.edu.au/~dersch/cbcorrect/cb.html. The program is very 
raw, but it may be useful to you and the open software panorama 
pictures community. It take two pictures,  and produce a third one that 

PTmender update

Hi everybody,

Just a quick update. I am "almost" done with the colour-brightness
adjustment. It is "almost" because it is still untested. Unfortunately
I don't have reliable Internet access, but I'll try to do a CVS commit
as soon as possible (I fixed a couple of bugs this week). So far
PTmender can stitch TIFF_m, but it can only generate TIFF.

The colour-brightness code is almost complete (I should it end today)
and I'll start testing. 

Please, one thing I really need is for somebody to run the example
from Helmut's web site
(http://www.path.unimelb.edu.au/~dersch/cbcorrect/cb.html) and send me
the debug.txt (I don't have a machine that can run PTStitcher). That
way I can compare my output with his output.

With respect to the algorithm used I can roughly describe it as
follows (of course this is my interpretation):

Compute all the histograms of all the intersections of photos, n *
(n-1)

Start with reference photo as corrected.

While there are photos to correct
  Find new photo with the largest overlap with corrected photos
    Compute Histogram of overlap of new photo with corrected photos
    Compute a curve that will match the photo to correct histogram to

Re: PTmender update

Hi everybody,

> Please, one thing I really need is for somebody to run the example
> from Helmut's web site
> (http://www.path.unimelb.edu.au/~dersch/cbcorrect/cb.html) and send me
> the debug.txt (I don't have a machine that can run PTStitcher). That
> way I can compare my output with his output.

I have tried, but the PTStitcher version available at that site did't 
produce any debug.txt. Is there a hidden parameter to enable this?

> With respect to the algorithm used I can roughly describe it as
> follows (of course this is my interpretation):
> 
> Compute all the histograms of all the intersections of photos, n *
> (n-1)
> 
> Start with reference photo as corrected.
> 
> While there are photos to correct
>   Find new photo with the largest overlap with corrected photos
>     Compute Histogram of overlap of new photo with corrected photos
>     Compute a curve that will match the photo to correct histogram to
>      the correctedPhotosHistogram 

This is the interesting part. I'm looking forward to see this code.

ciao
   Pablo

Re: PTmender update

 Pablo d'Angelo twisted the bytes to say:

 Pablo> Hi everybody,
 >> Please, one thing I really need is for somebody to run the example
 >> from Helmut's web site
 >> (http://www.path.unimelb.edu.au/~dersch/cbcorrect/cb.html) and send me
 >> the debug.txt (I don't have a machine that can run PTStitcher). That
 >> way I can compare my output with his output.

 Pablo> I have tried, but the PTStitcher version available at that site did't 
 Pablo> produce any debug.txt. Is there a hidden parameter to enable this?

hi Pablo, 

The option is available by default in version 2.6. From what I have
seen it should generate the file by default every time the colour
correction is done. The file should be placed in the current directory.

--
Daniel M. German
http://turingmachine.org/
http://silvernegative.com/
dmg (at) uvic (dot) ca
replace (at) with  <at>  and (dot) with .

PTmender Version 0.2 "The enlightment"

Hi everybody,

I am happy to announce that colour correction is now in beta. It
works, at least, for Helmut's test.

2 points to take into consideration:

* Correction curves should be identical between PTstitcher and
  PTmender.

* The output histogram errors will not be the same, because the
  algorithm is not deterministic (it uses the rand() function, which
  results in different values between different architectures). I am
  very curious to see the Debug.txt file of the Helmut example when
  run in a Linux box. Can somebody try it please? 

Of course, bug reports are welcome.

--
daniel

--
Daniel M. German
http://turingmachine.org/
http://silvernegative.com/
dmg (at) uvic (dot) ca
replace (at) with  <at>  and (dot) with .

-------------------------------------------------------

Re: PTmender Version 0.2 "The enlightment"

Daniel M. German schrieb:
> Hi everybody,
> 
> I am happy to announce that colour correction is now in beta. It
> works, at least, for Helmut's test.

Great!

> * The output histogram errors will not be the same, because the
>   algorithm is not deterministic (it uses the rand() function, which
>   results in different values between different architectures). I am
>   very curious to see the Debug.txt file of the Helmut example when
>   run in a Linux box. Can somebody try it please? 
> 
> Of course, bug reports are welcome.

I'll try and compare the results once the anonymous cvs on sf contains your
changes. (always takes a few hours..)

ciao
  Pablo

-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click

Bug in configure script?

Hi everybody,

I keep hacking at PTmender and last night (ok, it is more than last
night, given that I was in Madrid and tonight I am in Quebec City) I
completed PSD_nomask support (not yet in CVS).

I discovered that there is a problem with endianism detection in the
configure script (or lack of detection, I might add). In my iboook
(which is little endian) the library was compiled as big endian (see
PT_BIGENDIAN in filter.h).

This needs to be fixed before I can release the next version of
PTmender. How do we proceed to do it?

Also, I have noticed some messages about problems/restrictions in PSD
support in Panotools. Can someone describe what are the
problems/limitations?

--
Daniel M. German                  "Sooner or later all the peoples of the world,
                                   without regard to the political system
                                   under which they live,
                                   will have to discover a way
   Martin Luther King, jr. ->      to live together in peace."
http://turingmachine.org/
http://silvernegative.com/
dmg (at) uvic (dot) ca
replace (at) with  <at>  and (dot) with .

Re: Bug in configure script?

Hi there,

--- "Daniel M. German" <dmgerman@...> wrote:

> I discovered that there is a problem with endianism detection in the
> configure script (or lack of detection, I might add). In my iboook
> (which is little endian) the library was compiled as big endian (see
> PT_BIGENDIAN in filter.h).
> 
> This needs to be fixed before I can release the next version of
> PTmender. How do we proceed to do it?
> 

I have committed an endian check to cvs. No access at the moment to an OSX machine, so if you
could please check it... :)

regards,
Doug

		
___________________________________________________________ 
Win a BlackBerry device from O2 with Yahoo!. Enter now. http://www.yahoo.co.uk/blackberry

-------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
for problems?  Stop!  Download the new AJAX search engine that makes
searching your log files as easy as surfing the  web.  DOWNLOAD SPLUNK!
http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click

Re: Bug in configure script?

Doug, that was quick! 

I have tested your changes. The library builds properly and the
endianism is correct for my ibook. 

Thanks a lot!

daniel

 Doug> Hi there,
 Doug> --- "Daniel M. German" <dmgerman@...> wrote:

 >> I discovered that there is a problem with endianism detection in the
 >> configure script (or lack of detection, I might add). In my iboook
 >> (which is little endian) the library was compiled as big endian (see
 >> PT_BIGENDIAN in filter.h).
 >> 
 >> This needs to be fixed before I can release the next version of
 >> PTmender. How do we proceed to do it?
 >> 

 Doug> I have committed an endian check to cvs. No access at the moment to an OSX machine, so if you
 Doug> could please check it... :)

 Doug> regards,
 Doug> Doug

--
Daniel M. German

PTmender version 0.3 "The mask"

A new version of PTmender is already in CVS.

The changes are mainly support for Photoshop PSD format, including
the creation of layer masks. 

There are two caveats, however: I have not implemented 16bit support
in masks yet, and there is no feathering support either.

In my opinion, PTmender is ready for real use.

I will probably take a break from development for a while. I need to
get back to take photos, after all :)

My plans for PTmender are:

In the short while, and in this order:

* Add support for flat output, including JPG (1 single layer)
* Add support for 16 bit creation of masks.
* Add feathering support.

In the longer term:

* Add support for 16 bit colour-brightness correction.

Also, PTmender hasn't been tested in correction of intensity only,
nor in correction of colours only (b and d options). It has only been
tested in the k option. Use at your own risk.