Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Jehan Pagès
Hi,

On Thu, Nov 13, 2014 at 11:04 PM, Elle Stone
 wrote:
> On 11/13/2014 09:51 AM, Jehan Pagès wrote:
>>>
>>> I take it as given that Pat, Jehan, and myself will do the actual work of
>>> assembling the images, yes? And all three of us will also contribute test
>>> images. We each have websites so we can put up pages with thumbnails of
>>> suggested test images for consideration.
>>
>>
>> Well for the "contribute test images" part, I'm not a photographer. I
>> have a good enough Canon camera and I can try to shoot some objects
>> with RAW on request though, but I won't guarantee the art quality for
>> these. Though you'll say that's not the purpose of this test set
>> anyway. :P
>
>
> Well, I think it's asking a lot to expect a test image to also be
> artistically pleasing, except for maybe for people's faces, and Pat David
> makes really good photographs of people's faces.
>
>> For the rest, yes, I can help organize, host if necessary, put up a
>> small website (nothing fancy, I'm not really into web UI design) or
>> something.
>
>
> If you are willing to put together a small website, that would be wonderful.
> Simple is better than complicated.

I'll do it soon.

> To allow for displaying and sorting through potential test images, I think
> jpeg image "large thumbnails" would suffice and would minimize demands on
> your server.
>
> For actual storage and distribution of a finished pack of test images plus
> associated raw files, server demands would go up. So that's a separate issue
> to be addressed.

I think I'll use TuxFamily which is my usual FLOSS community host when
I'm unsure my small server will handle correctly the load. So that
should not be a problem.

>> Also we should probably broaden the user list to encompass other FLOSS
>> graphics projects which would also be interested and can provide user
>> input and well as contribute photos. I don't think it should stay a
>> GIMP-only project.
>
>
> +1.

As soon as I finished the website, and as you validate the way the
project is presented (I'll do a small base text, but I'll expect
fixes), I'll try and reach out to several graphics project.
Also Steve Czajka from GIMP magazine said that if we can put up a
small text before the end of November, it should go into January
release without much problem (apparently GIMP magazine is getting
monthly from now on). So we should be able to gather interested people
this way, I hope. :-)

>> As for gathering user input for instance, if every potential
>> interested project (GIMP, ImageMagick, Krita, G'Mic... are the ones
>> I'd think without much searching) would make a blog post to ask user
>> input and later user raw photos once we got the input, we could have a
>> usable test image set in no times (which can anyway always improve
>> with time and more user input).
>
>
>> I think the first step would be a small simple page to present the
>> project with a way to contact us. We could have a basic survey there
>> "what are you looking for when you want test image and for what
>> purpose?". I can do such a thing.
>
>
> Sounds good to me.
>
> It might be nice to put up a page of sample test image "large thumbnails"
> comprising a mix of synthetic test images, colorful subjects, and faces. And
> we could include a list of links to existing useful but copyright-encumbered
> test images to give people an idea of the kinds of test images we'd like to
> put together.
>
> I can supply thumbnails of some of the synthetic and colorful test images
> that I've put together. Would "512px max dimension jpegs" be a good size?

I guess it is.

> The thumbnail images wouldn't represent any composite test images. But once
> we have a good set of images to work with, suitable composite test images
> would be easy to put together.
>
>> Then we can redirect people to this URI anytime, and ask projects to
>> do blog posts. Because staying email-only is a way to have the project
>> end up in limbo with time passing.
>>
>
>> Then at some point, we go to the "gathering step" where we ask users
>> to send raw photos, ask them to accept the licensing (I would agree
>> with CC by and CC by-sa personally), and ask them to get the explicit
>> agreement of the model (maybe with a written agreement scan even?).
>> And of course we could present what we already have.
>>
>> For the decisions on exact contents, I would let any of you in charge
>> though, since I am not the best fit to decide.
>
>
> That's a place where input from users would be really, really helpful.
> Otherwise the people putting together the test images don't know what other
> people would find useful.

Ok. :-)

Jehan

>
> Elle
>
> ___
> gimp-user-list mailing list
> List address:gimp-user-l...@gnome.org
> List membership: https://mail.gnome.org/mailman/listinfo/gimp-user-list
> List archives:   https://mail.gnome.org/archives/gimp-user-list
___
gimp-developer-list mailing list
List 

Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Elle Stone

On 11/13/2014 09:51 AM, Jehan Pagès wrote:

I take it as given that Pat, Jehan, and myself will do the actual work of
assembling the images, yes? And all three of us will also contribute test
images. We each have websites so we can put up pages with thumbnails of
suggested test images for consideration.


Well for the "contribute test images" part, I'm not a photographer. I
have a good enough Canon camera and I can try to shoot some objects
with RAW on request though, but I won't guarantee the art quality for
these. Though you'll say that's not the purpose of this test set
anyway. :P


Well, I think it's asking a lot to expect a test image to also be 
artistically pleasing, except for maybe for people's faces, and Pat 
David makes really good photographs of people's faces.



For the rest, yes, I can help organize, host if necessary, put up a
small website (nothing fancy, I'm not really into web UI design) or
something.


If you are willing to put together a small website, that would be 
wonderful. Simple is better than complicated.


To allow for displaying and sorting through potential test images, I 
think jpeg image "large thumbnails" would suffice and would minimize 
demands on your server.


For actual storage and distribution of a finished pack of test images 
plus associated raw files, server demands would go up. So that's a 
separate issue to be addressed.



Also we should probably broaden the user list to encompass other FLOSS
graphics projects which would also be interested and can provide user
input and well as contribute photos. I don't think it should stay a
GIMP-only project.


+1.


As for gathering user input for instance, if every potential
interested project (GIMP, ImageMagick, Krita, G'Mic... are the ones
I'd think without much searching) would make a blog post to ask user
input and later user raw photos once we got the input, we could have a
usable test image set in no times (which can anyway always improve
with time and more user input).



I think the first step would be a small simple page to present the
project with a way to contact us. We could have a basic survey there
"what are you looking for when you want test image and for what
purpose?". I can do such a thing.


Sounds good to me.

It might be nice to put up a page of sample test image "large 
thumbnails" comprising a mix of synthetic test images, colorful 
subjects, and faces. And we could include a list of links to existing 
useful but copyright-encumbered test images to give people an idea of 
the kinds of test images we'd like to put together.


I can supply thumbnails of some of the synthetic and colorful test 
images that I've put together. Would "512px max dimension jpegs" be a 
good size?


The thumbnail images wouldn't represent any composite test images. But 
once we have a good set of images to work with, suitable composite test 
images would be easy to put together.



Then we can redirect people to this URI anytime, and ask projects to
do blog posts. Because staying email-only is a way to have the project
end up in limbo with time passing.




Then at some point, we go to the "gathering step" where we ask users
to send raw photos, ask them to accept the licensing (I would agree
with CC by and CC by-sa personally), and ask them to get the explicit
agreement of the model (maybe with a written agreement scan even?).
And of course we could present what we already have.

For the decisions on exact contents, I would let any of you in charge
though, since I am not the best fit to decide.


That's a place where input from users would be really, really helpful. 
Otherwise the people putting together the test images don't know what 
other people would find useful.


Elle
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Elle Stone

On 11/13/2014 10:02 AM, Simon Budig wrote:


I do have access to a x-rite color checker which should be in a
reasonably good condition (it is slightly older but has been stored in
the dark).

I could try to set up a colorful scene and shoot it in RAW with my
Nikon D40. Would this be of any help?


Yes. A colorchecker is very often found in composite test images. Given 
the number of websites that post pictures of colorchecker charts, 
hopefully there are no copyright restrictions on photographs of same.




(I am right now unsure about the correct lighting, there *should* be
some daylight neon tubes around, but I am at the moment unsure where
they are...)


You raise a good point. On the one hand, natural daylight on a sunny day 
(D65ish) and direct morning and late afternoon/early evening sunlight 
(D50ish) are wonderful full spectrum light sources that really bring out 
colors. On the other hand, studio lighting is either iffy or expensive 
and probably sometimes both (my own studio lighting is on the "cheap and 
iffy" side).


Where I am sunlight is in short supply until spring, so it's studio 
lighting or nothing. For flourescent and LED lighting, the fact that the 
label says "daylight" is not very informative by itself. The other half 
of the story is the CRI, "color rendering index" 
(http://en.wikipedia.org/wiki/Color_rendering_index). Anything 92 and 
above is probably good enough.


Elle
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Gary Aitken
On 11/11/14 17:32, Chris Mohler wrote:

> The 2002 version contains a *wide* variety of things that can go
> spectacularly wrong when printing (eg for one, over saturation of
> black completely destroys the image of the watch), and was a real life
> saver when I was setting up the workflow of: Workstation -> Digital
> RIP -> Large Format Printer.  It's a beastly image to try and print
> correctly, and I imagine the 2009 version is just as devious :)

For the curious and not overly competent, can you shed some light on 
the kinds of issues you had and how you dealt with them?  
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Simon Budig
Jehan Pagès (jehan.marmott...@gmail.com) wrote:
> Well for the "contribute test images" part, I'm not a photographer. I
> have a good enough Canon camera and I can try to shoot some objects
> with RAW on request though, but I won't guarantee the art quality for
> these. Though you'll say that's not the purpose of this test set
> anyway. :P

I do have access to a x-rite color checker which should be in a
reasonably good condition (it is slightly older but has been stored in
the dark).

I could try to set up a colorful scene and shoot it in RAW with my
Nikon D40. Would this be of any help?

(I am right now unsure about the correct lighting, there *should* be
some daylight neon tubes around, but I am at the moment unsure where
they are...)

Bye,
Simon

-- 
  si...@budig.de  http://simon.budig.de/
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Jehan Pagès
Hi,

On Thu, Nov 13, 2014 at 1:55 PM, Elle Stone
 wrote:
> On 11/12/2014 08:32 PM, Pat David wrote:
>>
>> Jehan,
>>
>> Ah, I did mean to link something, but thank you for linking my Flickr
>> account.  Note that for the most part, I have the RAW files from any of my
>> photos that I'd be happy to release for whatever purposes are needed.
>>
>> I also have model releases for all of my images as well, and can include
>> them as needed (there might need to be some discussion surrounding how
>> that
>> release is made available and in what context as they contain probably
>> sensitive information regarding the models themselves).
>>
>> To Elle's point about licensing, I can make them CC-BY(-SA) as
>> appropriate.
>>
>> To Elle's other point about color, I would imagine that the most desirable
>> solution would be to host the actual RAW files themselves (hopefully from
>> a
>> variety of capture sources/cameras).  Any further work around conversion
>> should probably be noted as needed to effectively reproduce the results.
>>
>> More importantly, if there is a consensus reached about certain types of
>> test images, I am also available to shoot new ones specifically for the
>> purpose.  I can rent other cameras/bodies as needed to support multiple
>> capture sources.  Please don't hesitate to ask if there's something that
>> may prove useful (I may not be coding, but perhaps I can at least help out
>> in this way).
>>
>
> It looks like we've reached the point of "let's do this", which will require
> assembling the actual images and ironing out a lot of details.
>
> I take it as given that Pat, Jehan, and myself will do the actual work of
> assembling the images, yes? And all three of us will also contribute test
> images. We each have websites so we can put up pages with thumbnails of
> suggested test images for consideration.

Well for the "contribute test images" part, I'm not a photographer. I
have a good enough Canon camera and I can try to shoot some objects
with RAW on request though, but I won't guarantee the art quality for
these. Though you'll say that's not the purpose of this test set
anyway. :P

For the rest, yes, I can help organize, host if necessary, put up a
small website (nothing fancy, I'm not really into web UI design) or
something.
Also we should probably broaden the user list to encompass other FLOSS
graphics projects which would also be interested and can provide user
input and well as contribute photos. I don't think it should stay a
GIMP-only project.
As for gathering user input for instance, if every potential
interested project (GIMP, ImageMagick, Krita, G'Mic... are the ones
I'd think without much searching) would make a blog post to ask user
input and later user raw photos once we got the input, we could have a
usable test image set in no times (which can anyway always improve
with time and more user input).

> Nicolas Robidoux has already pointed to a good collection of images, which
> also provides a model for including all the required copyright and image
> generation information.
>
> Contributions of proposed test images from other people are of course very,
> very welcome.
>
> Would it be appropriate to perhaps ascertain who all wants to be actively
> involved in putting together a pack of test images? And then perhaps take
> this discussion off-list until we have something concrete to present for
> further input from GIMP users?

I think the first step would be a small simple page to present the
project with a way to contact us. We could have a basic survey there
"what are you looking for when you want test image and for what
purpose?". I can do such a thing.
Then we can redirect people to this URI anytime, and ask projects to
do blog posts. Because staying email-only is a way to have the project
end up in limbo with time passing.

Then at some point, we go to the "gathering step" where we ask users
to send raw photos, ask them to accept the licensing (I would agree
with CC by and CC by-sa personally), and ask them to get the explicit
agreement of the model (maybe with a written agreement scan even?).
And of course we could present what we already have.

For the decisions on exact contents, I would let any of you in charge
though, since I am not the best fit to decide. My main purpose here is
to improve Free Software and Graphics environment. :-)

Jehan

> Best regards,
> Elle Stone
>
> ___
> gimp-user-list mailing list
> List address:gimp-user-l...@gnome.org
> List membership: https://mail.gnome.org/mailman/listinfo/gimp-user-list
> List archives:   https://mail.gnome.org/archives/gimp-user-list
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Elle Stone

On 11/13/2014 09:23 AM, Jehan Pagès wrote:

Otherwise I don't think we are in a hurry, and the set can also be
gathered slowly with time and patience

+1
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Jehan Pagès
Hi,

On Thu, Nov 13, 2014 at 2:32 AM, Pat David  wrote:
> Jehan,
>
> Ah, I did mean to link something, but thank you for linking my Flickr
> account.  Note that for the most part, I have the RAW files from any of my
> photos that I'd be happy to release for whatever purposes are needed.
>
> I also have model releases for all of my images as well, and can include
> them as needed (there might need to be some discussion surrounding how that
> release is made available and in what context as they contain probably
> sensitive information regarding the models themselves).

Yes I wanted to make a point about this as well, and I forgot. Thanks
for reminding. On the legal point of view, I think we will need to
have:
1/ the explicit agreement of the photographer (licensing);
2/ the explicit agreement of the model (or their legal representative,
i.e. parents usually when they are underage) if there are human being
shown and recognizable in the photograph.

This way, we make sure to have photos usable under any circumstances.
That's more bothering, but better be safe than sorry.

> To Elle's point about licensing, I can make them CC-BY(-SA) as appropriate.
>
> To Elle's other point about color, I would imagine that the most desirable
> solution would be to host the actual RAW files themselves (hopefully from a
> variety of capture sources/cameras).  Any further work around conversion
> should probably be noted as needed to effectively reproduce the results.
>
> More importantly, if there is a consensus reached about certain types of
> test images, I am also available to shoot new ones specifically for the
> purpose.  I can rent other cameras/bodies as needed to support multiple
> capture sources.  Please don't hesitate to ask if there's something that
> may prove useful (I may not be coding, but perhaps I can at least help out
> in this way).

Well if that costs you money (renting camera/model) and is done for
this specific task, I would be for crowdfunding a bit. Unless that's
not your sole purpose and you are planning to use these photos somehow
else.
Otherwise I don't think we are in a hurry, and the set can also be
gathered slowly with time and patience (when a photograph hires a
model or is hired for some usual shooting purpose, one can just ask
the model if he'd accept some photos, and which ones, to be used in a
test set).

Jehan

> --
> pat david
> http://blog.patdavid.net
> ___
> gimp-user-list mailing list
> List address:gimp-user-l...@gnome.org
> List membership: https://mail.gnome.org/mailman/listinfo/gimp-user-list
> List archives:   https://mail.gnome.org/archives/gimp-user-list
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-13 Thread Elle Stone

On 11/12/2014 08:32 PM, Pat David wrote:

Jehan,

Ah, I did mean to link something, but thank you for linking my Flickr
account.  Note that for the most part, I have the RAW files from any of my
photos that I'd be happy to release for whatever purposes are needed.

I also have model releases for all of my images as well, and can include
them as needed (there might need to be some discussion surrounding how that
release is made available and in what context as they contain probably
sensitive information regarding the models themselves).

​To Elle's point about licensing, I can make them CC-BY(-SA) as appropriate.

To Elle's other point about color, I would imagine that the most desirable
solution would be to host the actual RAW files themselves (hopefully from a
variety of capture sources/cameras).  Any further work around conversion
should probably be noted as needed to effectively reproduce the results.

More importantly, if there is a consensus reached about certain types of
test images, I am also available to shoot new ones specifically for the
purpose.  I can rent other cameras/bodies as needed to support multiple
capture sources.  Please don't hesitate to ask if there's something that
may prove useful (I may not be coding, but perhaps I can at least help out
in this way).  ​



It looks like we've reached the point of "let's do this", which will 
require assembling the actual images and ironing out a lot of details.


I take it as given that Pat, Jehan, and myself will do the actual work 
of assembling the images, yes? And all three of us will also contribute 
test images. We each have websites so we can put up pages with 
thumbnails of suggested test images for consideration.


Nicolas Robidoux has already pointed to a good collection of images, 
which also provides a model for including all the required copyright and 
image generation information.


Contributions of proposed test images from other people are of course 
very, very welcome.


Would it be appropriate to perhaps ascertain who all wants to be 
actively involved in putting together a pack of test images? And then 
perhaps take this discussion off-list until we have something concrete 
to present for further input from GIMP users?


Best regards,
Elle Stone
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-12 Thread Pat David
Jehan,

Ah, I did mean to link something, but thank you for linking my Flickr
account.  Note that for the most part, I have the RAW files from any of my
photos that I'd be happy to release for whatever purposes are needed.

I also have model releases for all of my images as well, and can include
them as needed (there might need to be some discussion surrounding how that
release is made available and in what context as they contain probably
sensitive information regarding the models themselves).

​To Elle's point about licensing, I can make them CC-BY(-SA) as appropriate.

To Elle's other point about color, I would imagine that the most desirable
solution would be to host the actual RAW files themselves (hopefully from a
variety of capture sources/cameras).  Any further work around conversion
should probably be noted as needed to effectively reproduce the results.

More importantly, if there is a consensus reached about certain types of
test images, I am also available to shoot new ones specifically for the
purpose.  I can rent other cameras/bodies as needed to support multiple
capture sources.  Please don't hesitate to ask if there's something that
may prove useful (I may not be coding, but perhaps I can at least help out
in this way).  ​



-- 
pat david
http://blog.patdavid.net
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-12 Thread Elle Stone

On 11/11/2014 07:32 PM, Chris Mohler wrote:

The license wording is a little ambiguous, but I interpret it to mean
that it's OK to use it for testing purposes, as long as you aren't
selling or otherwise distributing the image itself.  I'm not sure how
that would relate to a project like GIMP, and it appears we'd need a
German speaker to even ask.


Keeping in mind that I am not a lawyer and corrections/additions to 
what's below are very welcome . . .


The license for the excellent (even if not quite "family-rated") Pixl 
test image reads like many/most test image licenses. The image can't be 
redistributed as part of a package of test images. Nor can it be used 
commercially.


Which raises the question of how to license a package of test images, 
hopefully in a way to be compatible with distribution alongside 
free/libre software.


CC-BY-SA (used by Wikipedia and compatible with free software) allows 
commercial use of images. AFAIK commercial use does require a release 
from all recognizable people in the image.


Releasing an image as public domain/CC0 (also used by Wikipedia and 
compatible with free software) requires consent of the person taking the 
photograph and also would seem to require the consent of any 
recognizable person being photographed.


Also, if there is a recognizable person in the image, the thorny issue 
of personality rights is raised 
(http://en.wikipedia.org/wiki/Personality_rights).


So two practical questions are how to license and how to distribute the 
test images.


Elle
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-11 Thread Chris Mohler
On Tue, Nov 11, 2014 at 9:15 AM, Elle Stone
 wrote:
> *What copyrighted or otherwise test images GIMP users might have downloaded
> from the internet and found useful, and
> *what kinds of test images GIMP users might have already put together,
> *for what particular testing purposes.

I have used this image for a wide variety of testing, mostly to do
with printing:
http://www.pixl.dk/download/ (first link)

(note to prudes, there is a woman's bare bottom included ;)

The license wording is a little ambiguous, but I interpret it to mean
that it's OK to use it for testing purposes, as long as you aren't
selling or otherwise distributing the image itself.  I'm not sure how
that would relate to a project like GIMP, and it appears we'd need a
German speaker to even ask.

The 2002 version contains a *wide* variety of things that can go
spectacularly wrong when printing (eg for one, over saturation of
black completely destroys the image of the watch), and was a real life
saver when I was setting up the workflow of: Workstation -> Digital
RIP -> Large Format Printer.  It's a beastly image to try and print
correctly, and I imagine the 2009 version is just as devious :)

/goes back to munching popcorn and lurking on the color threads...

Chris
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-11 Thread Elle Stone

Here are links to some sample collections of copyrighted test images:

http://www.pcstats.com/articleimages/200601/samsungspp2040_300dpi1.jpg
http://www.northlight-images.co.uk/article_pages/test_images.html
http://www.outbackphoto.com/printinginsights/pi048/essay.html



So the first question is: What kind of test images, for what kinds of
testing, do you all, as a diverse group of GIMP users and developers, wish
you had access to?

Elle


I know we have a diverse and talented group of GIMP users. It seems to 
me that assembling a good set of test images really does need to start 
with feedback from GIMP users as to what kind of test images they might 
find useful.


I already know which test images I've downloaded from the internet over 
the years, for what purposes. I also know which test images I've put 
together, and for what purposes. It could be that no one else would find 
these images useful.


So the important question is:

*What copyrighted or otherwise test images GIMP users might have 
downloaded from the internet and found useful, and

*what kinds of test images GIMP users might have already put together,
*for what particular testing purposes.

It could be the case that hardly any GIMP user has ever used any test 
images (doubtful).


It could be the case that the phrase "test image" needs further 
clarification. Here are some random examples of when an appropriate test 
image might be useful:


1. Somewhere in my collection of copyrighted test images that can't be 
shared, there is a brightly colored montage that includes buildings 
along a stretch of beach, a parrot, and a woman's face. I found that 
test image very useful for learning about various "conversion to black 
and white" routines.


2. Grayscale block and grayscale and color gradient images are useful 
for testing things like:


*How "LCMS2 plus the monitor+monitor-profile plus the GIMP color 
management settings" affects how dark an image can get before dark gray 
and black can't be visually distinguished.
*How smoothly the monitor+profile displays and the printer+profile 
prints smooth gradients.


3. A granger rainbow can be used for all kinds of testing, such as 
"Visually what kind of overlap is there between any given RGB working 
space and the monitor+profile" and "What happens when extremely 
saturated colors from a selected RGB working space are converted to the 
printer profile using perceptual intent".


So before getting too far down the road of randomly assembling images 
and hoping someone finds them useful enough to justify the time and 
effort that would be involved, I think it would be helpful, actually 
essential, to have feedback from at least a few GIMP users as to what 
kinds of test images you've used in the past (preferably with links to 
example test images), for what specific testing purposes.


Best regards,
Elle Stone
--
http://ninedegreesbelow.com
Color management and free/libre photography
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-11 Thread Jehan Pagès
Hi,

On Mon, Nov 10, 2014 at 5:30 PM, Pat David  wrote:
> I can't add much to the color discussion, but at least I can offer up just
> about any of my images as RAW for any testing (just about everything I
> shoot is cc-by-sa usually).  If anyone finds one of mine they would like,
> just let me know and I'd be happy to provide the camera raw file.

I was actually gonna suggest to perhaps ask you, in particular for the
"nice set of pictures of people's faces, young to old, male and
female, of diverse skin colors." since I know you like to shoot
people. That may be a start.
Since you forgot to add a link to your photos (maybe out of
humility?), here they are: https://www.flickr.com/photos/patdavid

Unfortunately I am about the same for the actual color discussion
part, and I'm not sure which photos are the best for
color/printer/algorithms/other tests, but I can for instance host the
test suite photos. Maybe under testimages.libreart.info or some
similar URL. So I suggest that if someone on this thread thinks that
this or that photo is interesting, tell me, and I'll make a test page
gathering raw photos. With time, I think we should be able to gather a
good suite of standard copyleft test photos.

Jehan

> On Fri, Nov 7, 2014 at 1:24 PM, Elle Stone 
> wrote:
>
>> On 11/07/2014 10:25 AM, Jehan Pagès wrote:
>>
>>> >A good test suite of "copyleft" images would be a nice thing to have,
 >whether for testing new and existing editing algorithms, or for whatever
 >testing that individual GIMP users might want to do on their own
 >(printer-related, for example).

>>> That's a good idea. What kind of images would be the most interesting?
>>> Basically should that be images, taken with a good digital camera, of
>>> a lot of objects of various colors?
>>> It could also be images with color gradients, I guess (sunset/rise and
>>> such)?
>>> Or do you already have such copyleft images at your disposal that you
>>> could provide?
>>> Indeed if we could gather these for access to anyone as reference
>>> (then various software could use them for their own tests), it would
>>> be great.
>>>
>>> Jehan
>>>
>>>
>> Nicolas Robidous's test image collection is very nice, in particular the
>> baby's face and the brightly colored buildings make great test images. His
>> images are already converted to sRGB, which means they can't really fully
>> exercise the color gamuts of reasonably decent printers and wider gamut
>> displays.
>>
>> I can make available "straight from the camera interpolated raw file, no
>> enhancements added" images of very saturated (outside the sRGB color gamut)
>> natural objects, mostly flowers.
>>
>> I've also put together various artificial color ramps, granger rainbows,
>> stepped gray scales, and such. And I have IT8 target shots from several
>> cameras, which I think the photographers would release under an appropriate
>> license.
>>
>> I wish that I had a nice set of pictures of people's faces, young to old,
>> male and female, of diverse skin colors. Skin tones are something that
>> everyone wants to get "just right", so faces make great test images. Such
>> photographs ideally would be shot raw under natural daylight, more or less
>> full frame, and properly white balanced, preferably with a white balancing
>> object discretely placed somewhere in the image frame (styrofoam cups, PVC
>> plastic, white coffee filters all work really well, often as well as
>> commercially available white balancing aids).
>>
>> High quality images with good gradients would be a nice addition to a
>> collection of test images. Interpolated raw files that have been output in
>> a wider gamut color space would be more versatile than images that have
>> already been converted to sRGB.
>>
>> Here are links to some sample collections of copyrighted test images:
>>
>> http://www.pcstats.com/articleimages/200601/samsungspp2040_300dpi1.jpg
>>
>> I would love to have enough copyleft images to put together a copyleft
>> composite similar to the one in the above link.
>>
>> http://www.northlight-images.co.uk/article_pages/test_images.html
>>
>> http://www.outbackphoto.com/printinginsights/pi048/essay.html
>>
>>
>> Thinking more about "what kind of images", it depends on who's testing
>> what. Here are some possible reasons for wanting test images:
>>
>> * Testing scaling algorithms.
>>
>> * Testing ICC profile conversions from wider gamut color spaces to printer
>> profiles and/or to display screen profiles.
>>
>> * Testing the quality of prints made by a commercial or personally owned
>> printers.
>>
>> So the first question is: What kind of test images, for what kinds of
>> testing, do you all, as a diverse group of GIMP users and developers, wish
>> you had access to?
>>
>> Elle
>>
>>
>> ___
>> gimp-developer-list mailing list
>> List address:gimp-developer-list@gnome.org
>> List membership: https://mail.gnome.org/mailman/listinfo/gimp-
>> developer-list
>> List arch

Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-10 Thread Pat David
I can't add much to the color discussion, but at least I can offer up just
about any of my images as RAW for any testing (just about everything I
shoot is cc-by-sa usually).  If anyone finds one of mine they would like,
just let me know and I'd be happy to provide the camera raw file.

On Fri, Nov 7, 2014 at 1:24 PM, Elle Stone 
wrote:

> On 11/07/2014 10:25 AM, Jehan Pagès wrote:
>
>> >A good test suite of "copyleft" images would be a nice thing to have,
>>> >whether for testing new and existing editing algorithms, or for whatever
>>> >testing that individual GIMP users might want to do on their own
>>> >(printer-related, for example).
>>>
>> That's a good idea. What kind of images would be the most interesting?
>> Basically should that be images, taken with a good digital camera, of
>> a lot of objects of various colors?
>> It could also be images with color gradients, I guess (sunset/rise and
>> such)?
>> Or do you already have such copyleft images at your disposal that you
>> could provide?
>> Indeed if we could gather these for access to anyone as reference
>> (then various software could use them for their own tests), it would
>> be great.
>>
>> Jehan
>>
>>
> Nicolas Robidous's test image collection is very nice, in particular the
> baby's face and the brightly colored buildings make great test images. His
> images are already converted to sRGB, which means they can't really fully
> exercise the color gamuts of reasonably decent printers and wider gamut
> displays.
>
> I can make available "straight from the camera interpolated raw file, no
> enhancements added" images of very saturated (outside the sRGB color gamut)
> natural objects, mostly flowers.
>
> I've also put together various artificial color ramps, granger rainbows,
> stepped gray scales, and such. And I have IT8 target shots from several
> cameras, which I think the photographers would release under an appropriate
> license.
>
> I wish that I had a nice set of pictures of people's faces, young to old,
> male and female, of diverse skin colors. Skin tones are something that
> everyone wants to get "just right", so faces make great test images. Such
> photographs ideally would be shot raw under natural daylight, more or less
> full frame, and properly white balanced, preferably with a white balancing
> object discretely placed somewhere in the image frame (styrofoam cups, PVC
> plastic, white coffee filters all work really well, often as well as
> commercially available white balancing aids).
>
> High quality images with good gradients would be a nice addition to a
> collection of test images. Interpolated raw files that have been output in
> a wider gamut color space would be more versatile than images that have
> already been converted to sRGB.
>
> Here are links to some sample collections of copyrighted test images:
>
> http://www.pcstats.com/articleimages/200601/samsungspp2040_300dpi1.jpg
>
> I would love to have enough copyleft images to put together a copyleft
> composite similar to the one in the above link.
>
> http://www.northlight-images.co.uk/article_pages/test_images.html
>
> http://www.outbackphoto.com/printinginsights/pi048/essay.html
>
>
> Thinking more about "what kind of images", it depends on who's testing
> what. Here are some possible reasons for wanting test images:
>
> * Testing scaling algorithms.
>
> * Testing ICC profile conversions from wider gamut color spaces to printer
> profiles and/or to display screen profiles.
>
> * Testing the quality of prints made by a commercial or personally owned
> printers.
>
> So the first question is: What kind of test images, for what kinds of
> testing, do you all, as a diverse group of GIMP users and developers, wish
> you had access to?
>
> Elle
>
>
> ___
> gimp-developer-list mailing list
> List address:gimp-developer-list@gnome.org
> List membership: https://mail.gnome.org/mailman/listinfo/gimp-
> developer-list
> List archives:   https://mail.gnome.org/archives/gimp-developer-list
>



-- 
pat david
http://blog.patdavid.net
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-07 Thread Elle Stone

On 11/07/2014 10:25 AM, Jehan Pagès wrote:

>A good test suite of "copyleft" images would be a nice thing to have,
>whether for testing new and existing editing algorithms, or for whatever
>testing that individual GIMP users might want to do on their own
>(printer-related, for example).

That's a good idea. What kind of images would be the most interesting?
Basically should that be images, taken with a good digital camera, of
a lot of objects of various colors?
It could also be images with color gradients, I guess (sunset/rise and such)?
Or do you already have such copyleft images at your disposal that you
could provide?
Indeed if we could gather these for access to anyone as reference
(then various software could use them for their own tests), it would
be great.

Jehan



Nicolas Robidous's test image collection is very nice, in particular the 
baby's face and the brightly colored buildings make great test images. 
His images are already converted to sRGB, which means they can't really 
fully exercise the color gamuts of reasonably decent printers and wider 
gamut displays.


I can make available "straight from the camera interpolated raw file, no 
enhancements added" images of very saturated (outside the sRGB color 
gamut) natural objects, mostly flowers.


I've also put together various artificial color ramps, granger rainbows, 
stepped gray scales, and such. And I have IT8 target shots from several 
cameras, which I think the photographers would release under an 
appropriate license.


I wish that I had a nice set of pictures of people's faces, young to 
old, male and female, of diverse skin colors. Skin tones are something 
that everyone wants to get "just right", so faces make great test 
images. Such photographs ideally would be shot raw under natural 
daylight, more or less full frame, and properly white balanced, 
preferably with a white balancing object discretely placed somewhere in 
the image frame (styrofoam cups, PVC plastic, white coffee filters all 
work really well, often as well as commercially available white 
balancing aids).


High quality images with good gradients would be a nice addition to a 
collection of test images. Interpolated raw files that have been output 
in a wider gamut color space would be more versatile than images that 
have already been converted to sRGB.


Here are links to some sample collections of copyrighted test images:

http://www.pcstats.com/articleimages/200601/samsungspp2040_300dpi1.jpg

I would love to have enough copyleft images to put together a copyleft 
composite similar to the one in the above link.


http://www.northlight-images.co.uk/article_pages/test_images.html

http://www.outbackphoto.com/printinginsights/pi048/essay.html


Thinking more about "what kind of images", it depends on who's testing 
what. Here are some possible reasons for wanting test images:


* Testing scaling algorithms.

* Testing ICC profile conversions from wider gamut color spaces to 
printer profiles and/or to display screen profiles.


* Testing the quality of prints made by a commercial or personally owned 
printers.


So the first question is: What kind of test images, for what kinds of 
testing, do you all, as a diverse group of GIMP users and developers, 
wish you had access to?


Elle

___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-07 Thread Nicolas Robidoux
I suggested
http://www.imagemagick.org/download/image-bank/16bit840x840images/README to
Elle.

Some details RE: how they were made are found in individual .txt files,
e.g.
http://www.imagemagick.org/download/image-bank/16bit840x840images/wave.txt
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list


Re: [Gimp-developer] [Gimp-user] Test images and test suite (was Re: GIMP should fork babl and GEGL)

2014-11-07 Thread Jehan Pagès
Hi,

On Fri, Nov 7, 2014 at 1:07 PM, Elle Stone
 wrote:
> On 11/06/2014 11:24 AM, Gary Aitken wrote:
>>
>> Even if the result is not obvious visually, I need a heads-up
>> so I can pay close attention to what's happening and undo the
>> operation if appropriate.
>
>
> On 11/07/2014 04:25 AM, Simos Xenitellis wrote:
>>
>> Is there a test suite available that could show the expected behavior?
>> If not, let's try to build one, both the test-suite code plus the images
>> (before/after).
>
>
> As Gary hinted, when doing "X" to an image, whatever "X" might be, sometimes
> results are visually obvious and sometimes not. Many times when editing
> images, I've done "X" and thought "X" was just fine, only to realize later
> that "X" had changed the file in an unwanted way that only became obvious
> several editing steps later.
>
> And sometimes "X" can make a change in an image that the monitor's limited
> color gamut can't display, that might only pop up visually when the image is
> printed or displayed on a wider gamut display.
>
> There are a lot of nice copyright-encumbered test images available on the
> internet, but not many "copyleft" test images, and not many images that
> reflect the full range of colors as captured by digital cameras and then
> interpreted using typical matrix input profiles.
>
> When I was testing unbounded sRGB image editing, the first task was to put
> together some appropriate test images and then figure out which tests needed
> to be made, as per Simos's "test suite".
>
> A good test suite of "copyleft" images would be a nice thing to have,
> whether for testing new and existing editing algorithms, or for whatever
> testing that individual GIMP users might want to do on their own
> (printer-related, for example).

That's a good idea. What kind of images would be the most interesting?
Basically should that be images, taken with a good digital camera, of
a lot of objects of various colors?
It could also be images with color gradients, I guess (sunset/rise and such)?
Or do you already have such copyleft images at your disposal that you
could provide?
Indeed if we could gather these for access to anyone as reference
(then various software could use them for their own tests), it would
be great.

Jehan

> Best regards,
> Elle Stone
> ___
> gimp-user-list mailing list
> List address:gimp-user-l...@gnome.org
> List membership: https://mail.gnome.org/mailman/listinfo/gimp-user-list
> List archives:   https://mail.gnome.org/archives/gimp-user-list
___
gimp-developer-list mailing list
List address:gimp-developer-list@gnome.org
List membership: https://mail.gnome.org/mailman/listinfo/gimp-developer-list
List archives:   https://mail.gnome.org/archives/gimp-developer-list