NYT article

hdri
1

The web version of the New York Times has an article about HDRI at:

It is basic and general, but I was surprised that it does not mention
Photosphere at all.
(maybe because there's no Windows version?)
anyway, I thought you might be interested, and perhaps someone more able
than me can write them to correct/expand the info?
regards,
Santiago

2

Hi Santiago,

I did see this article, thanks. I was annoyed that they didn't mention Photosphere or even HDRshop, but then neither are really commercial programs, and the article was clearly targeted at a more mainstream audience. Judy Lai is in fact working on documenting Photosphere, and I have another friend working on a port for Windows and Linux. So, maybe it will be commercial, one day! In any case, I plan to continue to distribute a free version for the Mac, since it's a basic bias of mine...

How are things with you? Do you think you'll make it to the next Radiance workshop in September?
-Greg

···

From: "Santiago Torres" <[email protected]>
Date: August 6, 2006 11:04:22 PM PDT

The web version of the New York Times has an article about HDRI at:

http://www.nytimes.com/2006/08/03/technology/03basics.html

It is basic and general, but I was surprised that it does not mention Photosphere at all.
(maybe because there's no Windows version?)
anyway, I thought you might be interested, and perhaps someone more able than me can write them to correct/expand the info?
regards,
Santiago

3

Hi Greg,

Great news about Photosphere (especially about the future free versions:)
And on the same subject, the last improvements were amazing. Thank you for
that. The false color images are great, I've been using them a lot. I just
wish there was a choice of a logarithmic scale.

As for me, I'm afraid I won't be able to get to this year's workshop. I
really am sorry since I wanted very much to get your tutorial on rtcontrib.
And the other titles seem very interesting also. And I'll miss meeting
everyone, for sure. I hope you all have a great time as usual.

Santiago

···

-----Original Message-----
From: [email protected]
[mailto:[email protected]]On Behalf Of Gregory J. Ward
Sent: Tuesday, August 08, 2006 4:32 AM
To: High Dynamic Range Imaging
Subject: Re: [HDRI] NYT article

Hi Santiago,

I did see this article, thanks. I was annoyed that they didn't
mention Photosphere or even HDRshop, but then neither are really
commercial programs, and the article was clearly targeted at a more
mainstream audience. Judy Lai is in fact working on documenting
Photosphere, and I have another friend working on a port for Windows
and Linux. So, maybe it will be commercial, one day! In any case, I
plan to continue to distribute a free version for the Mac, since it's
a basic bias of mine...

How are things with you? Do you think you'll make it to the next
Radiance workshop in September?
-Greg

> From: "Santiago Torres" <[email protected]>
> Date: August 6, 2006 11:04:22 PM PDT
>
> The web version of the New York Times has an article about HDRI at:
>
> http://www.nytimes.com/2006/08/03/technology/03basics.html
>
> It is basic and general, but I was surprised that it does not
> mention Photosphere at all.
> (maybe because there's no Windows version?)
> anyway, I thought you might be interested, and perhaps someone more
> able than me can write them to correct/expand the info?
> regards,
> Santiago

_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

4

Hi Santiago,

The false color option in Photosphere does do logarithmic mappings -- just specify a non-zero lower limit, and the scale becomes logarithmic. (This is the default if the "Auto" box is checked -- will be spelled out in the documentation, eventually!)

Sorry we'll miss you in Leicester!
-Greg

5

Hi Santiago and All,

I've just had a look at this article, and the images are
quite obviously created with a program called Photomatix.
This program combines multiple exposures, and also has a
tone reproduction algorithm built-in.

Unfortunately, most people on Flickr using HDR are using
this program. My annoyance stems from the fact that the
tone mapping algorithm produces very unnatural results, and
is about the poorest algorithm that I know for dynamic range
reduction. It is very easy to recognize Photomatix output as
a result --- it looks too compressed with too much local
contrast.

The apologetic comments at the end of the NYC article about
the unnatural look of the Photomatix output makes no sense to
me. There is no good reason to offer such a poor tone mapper
as the only choice in any program.

As a result, the Flickr community is starting to equate HDR
with non-photorealistic rendering, which in my view is very
undesirable. The NYC article is not helping in that respect
either, as it does not present a balanced view of the state
of the art in HDR.

High dynamic range imaging is an exciting technology for many,
as it allows us to take pictures and do image processing that
we couldn't do before. However, I feel that Photomatix, in
combination with its apparent, but undeserved popularity on
Flickr, is doing more harm than good to help establish HDR as
a useful technology.

Right, that's me finished venting.

Cheers,
Erik

6

Hi Erik

           You took the words right out of my mouth, i said the same thing
in a Flickr group.

Tom.

···

----- Original Message -----
From: "E. Reinhard" <[email protected]>
To: "High Dynamic Range Imaging" <[email protected]>
Sent: Thursday, August 10, 2006 7:07 AM
Subject: [HDRI] Re: NYT article

Hi Santiago and All,

I've just had a look at this article, and the images are
quite obviously created with a program called Photomatix.
This program combines multiple exposures, and also has a
tone reproduction algorithm built-in.

Unfortunately, most people on Flickr using HDR are using
this program. My annoyance stems from the fact that the
tone mapping algorithm produces very unnatural results, and
is about the poorest algorithm that I know for dynamic range
reduction. It is very easy to recognize Photomatix output as
a result --- it looks too compressed with too much local
contrast.

The apologetic comments at the end of the NYC article about
the unnatural look of the Photomatix output makes no sense to
me. There is no good reason to offer such a poor tone mapper
as the only choice in any program.

As a result, the Flickr community is starting to equate HDR
with non-photorealistic rendering, which in my view is very
undesirable. The NYC article is not helping in that respect
either, as it does not present a balanced view of the state
of the art in HDR.

High dynamic range imaging is an exciting technology for many,
as it allows us to take pictures and do image processing that
we couldn't do before. However, I feel that Photomatix, in
combination with its apparent, but undeserved popularity on
Flickr, is doing more harm than good to help establish HDR as
a useful technology.

Right, that's me finished venting.

Cheers,
Erik

_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

7

Photomatix is really very limited - I think the gallery images they show are not representative of typical output one can expect. Obviously tone mapping is a key ingredient in using HDR images on standard displays and print media. I have found the tone mapping algorithms discussed in the text, and included on the accompanying DVD, of the HDRI book to be a great asset in exploring tone mapping. Anyone who is interested in tone mapping should really take the time to read the chapters in the book dedicated to this topic. I am especially intrigued by the "photographic" tone mapping algorithm - in fact it inspired me to read all three of Ansel Adams' classic works on his photographical philosophy (The Camera, The Negative and The Print). Implementing the Zone System in HDR tone-mapping is an interesting approach, especially with a basis on the neutral 17% gray card (Zone 5). From a Radiance standpoint, is this 17% gray card in the macbeth.cal file? It would be interesting to include it in renderings and use the Zone System with the photographic_tm tone mapping operator to explore various variations on the original, high-bit rendering output.

Thanks Eric!

kirk

···

--------------------------------------------------
Kirk L. Thibault, Ph.D.

Biomechanics, Inc.
439 S. Governor Printz Blvd.
Essington, PA 19029

p. 215.271.7720
f 215.271.7740

[email protected]
skype: kirkthibault

On Aug 10, 2006, at 7:07 AM, E. Reinhard wrote:

Hi Santiago and All,

I've just had a look at this article, and the images are
quite obviously created with a program called Photomatix.
This program combines multiple exposures, and also has a
tone reproduction algorithm built-in.

Unfortunately, most people on Flickr using HDR are using
this program. My annoyance stems from the fact that the
tone mapping algorithm produces very unnatural results, and
is about the poorest algorithm that I know for dynamic range
reduction. It is very easy to recognize Photomatix output as
a result --- it looks too compressed with too much local
contrast.

The apologetic comments at the end of the NYC article about
the unnatural look of the Photomatix output makes no sense to
me. There is no good reason to offer such a poor tone mapper
as the only choice in any program.

As a result, the Flickr community is starting to equate HDR
with non-photorealistic rendering, which in my view is very
undesirable. The NYC article is not helping in that respect
either, as it does not present a balanced view of the state
of the art in HDR.

High dynamic range imaging is an exciting technology for many,
as it allows us to take pictures and do image processing that
we couldn't do before. However, I feel that Photomatix, in
combination with its apparent, but undeserved popularity on
Flickr, is doing more harm than good to help establish HDR as
a useful technology.

Right, that's me finished venting.

Cheers,
Erik

_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

8

Hi Erik,

You already know my opinion of Flickr and the over-Photomatixed results, but I see this article and the exposure and popularity of this "effect" on Flickr as both an expected and not entirely negative thing. The basics are that in the greater world of photo hobbyists (which Flickr represents) no one knows *what* HDR is, aside from the fact the letters mean "High Dynamic Range" and that Adobe put it in Photoshop CS2 and some other companies sell software to do something similar. Keep in mind these same people also have *no idea* what all can be done with HDR.

I would guess that the vast majority of photographers (pro, hobbyist, and "holiday snappers") only know what Adobe said in their marketing of this "new feature in CS2," or what the adventurous ones read on Luminouslandscape.com, and that is roughly that "HDR images can contain a greater range of tones and colors than other computer images, but it cannot be displayed on your monitor or manipulated much at this time." And most likely these same folks have *not* been struggling with the lack of range in "regular" photography - pros carefully choose their lighting conditions, and hobbyists frequently just don't notice. So the idea that HDR capture and processing are useful for extending the visual range of an image ends up getting lost when these same consumers see "well-tonemapped" results that look not too different from the original(s) or not too different from other photos they've seen. It is a lot of work to get what most people (I've talked to anyway) see as a very small visual difference.

Plus, in a content-competitive social community like Flickr (or even the NYTimes) you have to do something dramatically different to stand out. Pushing all the sliders in Photomatix "to 11" to get a wild, painterly and oversaturated compression of dynamic range is certainly different, and when people ask "how'd you do that" they can say "with that cool new HDR thing." If all Photomatix output looked just like Photoshop's tonemapping, or just like a nice, well-exposed photo, very few would ask "how'd you do that" and HDR very well might not be mentioned in the New York Times at all. Again, I strongly believe that this is because the subtleties in a resulting "photographic" tonemapped image (as well as the greater impact on image-making and photography) is not as immediately noticeable to most people.

Finally, I think this "Flickr abuse" is somewhat typical of what happens when a "new" technology is first put into the hands of the mass-market; people naturally try to push all the buttons and see what lights up, and in the end they decide what they like and what becomes popular for a time. I've watched this same process repeatedly over 20 years of "3D images" being publicized, and even now I don't think that a shiny car in a white void is any better than a chrome sphere on tiles, and "3D" is just beginning to be used judiciously. Of course I don't think what's on Flickr "is HDR" by any stretch of the imagination, but it *is* the most visually distinct result of an HDR process that people can access today. The true aesthetics of the process and technology are some time from shaking themselves out (and the use of HDR in video games is just another example of this).

The only suggestion to alleviate your particular frustration with Flickr and Photomatix would be to write to the makers of Photomatix (and Photoshop, et al) and ask them to include some of the existing different tonemapping methods in their next release. Then at least your tools would also be in the hands of the people on Flickr.

-Mark

BTW, you might obviously note that my opinions are those of an image-creator interested in the application of HDRI, not a researcher committed to its technical development and understanding. I feel the same need to vent when something I've created is grossly misused or misunderstood.

···

On Aug 10, 2006, at 7:07 AM, E. Reinhard wrote:

As a result, the Flickr community is starting to equate HDR
with non-photorealistic rendering, which in my view is very
undesirable. The NYC article is not helping in that respect
either, as it does not present a balanced view of the state
of the art in HDR.

9

Hello Eric,

I also seldomly use the Photomatix algorithm for the same reason. I had several talks with the Author of the software about this and i think it should be mentioned that the current beta version includes a new global operator.
The operator is especially useful for panoramic use as it will do a "matching" map of several images.
I used this TMO for a MDR stitch of a panorama (by Wolfgang Stich) and the results are very pleasing:
http://www.austria-360.at/kaernten/page-gletschermuehle.html

(Sorry for this commercial break, but i thought, it should be done justice to the current development of the software)

Best regards
Bernhard

E. Reinhard schrieb:

···

My annoyance stems from the fact that the
tone mapping algorithm produces very unnatural results, and
is about the poorest algorithm that I know for dynamic range
reduction. It is very easy to recognize Photomatix output as
a result --- it looks too compressed with too much local
contrast.

10

That pano is very impressive.

···

--------------------------------------------------
Kirk L. Thibault, Ph.D.

Biomechanics, Inc.
439 S. Governor Printz Blvd.
Essington, PA 19029

p. 215.271.7720
f 215.271.7740

[email protected]
skype: kirkthibault

On Aug 10, 2006, at 4:41 PM, Bernhard Vogl wrote:

Hello Eric,

I also seldomly use the Photomatix algorithm for the same reason. I had several talks with the Author of the software about this and i think it should be mentioned that the current beta version includes a new global operator.
The operator is especially useful for panoramic use as it will do a "matching" map of several images.
I used this TMO for a MDR stitch of a panorama (by Wolfgang Stich) and the results are very pleasing:
http://www.austria-360.at/kaernten/page-gletschermuehle.html

(Sorry for this commercial break, but i thought, it should be done justice to the current development of the software)

Best regards
Bernhard

E. Reinhard schrieb:

My annoyance stems from the fact that the
tone mapping algorithm produces very unnatural results, and
is about the poorest algorithm that I know for dynamic range
reduction. It is very easy to recognize Photomatix output as
a result --- it looks too compressed with too much local
contrast.

_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

11

Hi Erik,

I would be interested to know whether your rating of the tone mapper offered in Photomatix as poor
is based on a comparative survey. If such a survey has been conducted, what were the criteria used
for rating the algorithms? What was the background and number of participants of the survey? And
which settings have been used for producing the Photomatix results?

Please note that only a fraction of the photos tagged with HDR at Flickr have been produced with
Photomatix. There are other software offering tone mapping capabilities, and of course Photoshop
CS2 is the most well-known and used by photographers. I believe you are grossly overestimating the
popularity of Photomatix. Furthermore, most of the Flickr's images produced with Photomatix have
been post-processed in other imaging software, and therefore are not the direct output of its tone
mapper.

Photomatix offers user controls to parametrize the algorithm to the user's personal taste, and many
at Flickr seem to like to dramatize the effect with settings that are far away from the default values.
The approach of Photomatix is to let the user decide what is best for his/her photos, and not have it
imposed by the engineer who designed the algorithm. Photoshop CS2 apparently follows the same
approach, as its "Local Adaptation" HDR Conversion method allows a great deal of control over the
mapping, better than with Photomatix in fact.

Why should a software be blamed for allowing photographers to adjust the settings and produce
tone mapped images that specialists in HDR imaging consider non-photorealistic?

Photomatix is intended for photographers and not for 3D artists or other categories of users.
Photographers in general do not wish to produce photorealistic images. They are not interested for
instance in adding motion blur to make their images look photorealistic, as motion blur is precisely
what photographers want to avoid. The same applies to HDR and tone mapping: most photographers
are not interested in taking several exposures for HDR if the final result is going to be a
"photorealistic" image that looks like a dull Low Dynamic Range photograph.

Geraldine Joffre

···

On Thu, 10 Aug 2006 12:07:10 +0100 (BST), E. Reinhard wrote

Hi Santiago and All,

I've just had a look at this article, and the images are
quite obviously created with a program called Photomatix.
This program combines multiple exposures, and also has a
tone reproduction algorithm built-in.

Unfortunately, most people on Flickr using HDR are using
this program. My annoyance stems from the fact that the
tone mapping algorithm produces very unnatural results, and
is about the poorest algorithm that I know for dynamic range
reduction. It is very easy to recognize Photomatix output as
a result --- it looks too compressed with too much local
contrast.

The apologetic comments at the end of the NYC article about
the unnatural look of the Photomatix output makes no sense to
me. There is no good reason to offer such a poor tone mapper
as the only choice in any program.

As a result, the Flickr community is starting to equate HDR
with non-photorealistic rendering, which in my view is very
undesirable. The NYC article is not helping in that respect
either, as it does not present a balanced view of the state
of the art in HDR.

High dynamic range imaging is an exciting technology for many,
as it allows us to take pictures and do image processing that
we couldn't do before. However, I feel that Photomatix, in
combination with its apparent, but undeserved popularity on
Flickr, is doing more harm than good to help establish HDR as
a useful technology.

Right, that's me finished venting.

Cheers,
Erik

_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

12

<snip>

Photographers in general do not wish to produce photorealistic images.
Geraldine Joffre

</snip>

This is a very interesting statement, especially in light of 3D modelers and renderers who are trying to produce "photorealistic" images by using HDR photographs as radiance maps and reflectance maps. Fascinating philosophical point of view! I guess it comes down to one's definition of what is "photorealistic".

kirk

···

On Aug 11, 2006, at 4:36 AM, Geraldine Joffre wrote:
--------------------------------------------------
Kirk L. Thibault, Ph.D.

Biomechanics, Inc.
439 S. Governor Printz Blvd.
Essington, PA 19029

p. 215.271.7720
f 215.271.7740

[email protected]
skype: kirkthibault

13

Also shows how out of touch they are. They will represent a hurdle that
everyone involved in evolving the HDR technology will have to deal with from
this point on, thank you Photomatix :wink:

Tom

···

----- Original Message -----
From: "Kirk Thibault" <[email protected]>
To: "High Dynamic Range Imaging" <[email protected]>
Sent: Friday, August 11, 2006 10:42 AM
Subject: Re: [HDRI] Re: NYT article

On Aug 11, 2006, at 4:36 AM, Geraldine Joffre wrote:
<snip>
>
> Photographers in general do not wish to produce photorealistic images.
> Geraldine Joffre
>
</snip>

This is a very interesting statement, especially in light of 3D
modelers and renderers who are trying to produce "photorealistic"
images by using HDR photographs as radiance maps and reflectance
maps. Fascinating philosophical point of view! I guess it comes
down to one's definition of what is "photorealistic".

kirk
--------------------------------------------------
Kirk L. Thibault, Ph.D.

Biomechanics, Inc.
439 S. Governor Printz Blvd.
Essington, PA 19029

p. 215.271.7720
f 215.271.7740

[email protected]
skype: kirkthibault

_______________________________________________
HDRI mailing list
[email protected]
http://www.radiance-online.org/mailman/listinfo/hdri

14

The quoted statement makes little sense to me as well, but perhaps for different reasons.

"Photorealism" is a term that (I thought) means something "looks like a [traditional] photograph." Of course, when anyone picks up a camera, they already have understood and accepted that the image it produces will (most likely) be a 2D, monocular view of the scene that is cropped by a rectangular sensor/ film plane, and subject to the technology of the capture method. The resulting image looks nothing like what a human would see at the time, but many people still enjoy photographs (!) because they communicate part of what was seen at the time by triggering the mind to have a reaction. Good photographers heighten the reaction to an image by choice of subject, framing, lighting, processing, etc. This is all a part of the subjective notion of "photorealism," and we see where this idea is stretched everyday in bleached out colors in car ads, heavily retouched fashion models, or even cropping a mother-in-law from a vacation photo.

What people have been trying to do since the inception of photography (the darkened room with a hole in one wall, or "camera obscura") is capture *some* of "what is seen by the eyes" at a specific time and place, and HDRI is trying to advance and redefine what is seen as a "traditional photograph" by capturing *more* of what is seen by the human eye (among other things). So, in my opinion, today's photographic HDRI (and tone-mapping) is quite "non-photorealistic." Isn't that the point? To reduce some of the limitations of photography?

Anyway, this is all "a storm in a teacup" to me. Every image that goes into or out of my computer is subject to my whims and the technology I have access to. I would imagine the same is true for most everyone else who takes photos - you post an image you personally think looks good. Plus, I've seen some truly awful results from other TMOs put into powerpoint presentations to show what is a "good result," but my idea of "awful" is just where the author's opinion departs from mine, as far as aesthetics are concerned. I don't see their example image or TMO as an obstacle to HDRI and tonemapping - next year (or next month) there will hopefully be a new advancement and maybe myself and many others will find a use for this new research as well... when it gets into our tools, that is.

-Mark

···

On Aug 11, 2006, at 10:42 AM, Kirk Thibault wrote:

On Aug 11, 2006, at 4:36 AM, Geraldine Joffre wrote:
<snip>

Photographers in general do not wish to produce photorealistic images.
Geraldine Joffre

</snip>

This is a very interesting statement, especially in light of 3D modelers and renderers who are trying to produce "photorealistic" images by using HDR photographs as radiance maps and reflectance maps. Fascinating philosophical point of view! I guess it comes down to one's definition of what is "photorealistic".

15

You are right. I should have made it clear in what sense I used the term "photorealistic" here. By
"photorealistic", I was referring to the attempt to mimic the effects of camera limitations in order to
make an image (a computer generated one for instance) look as if it had been taken with a camera.

If you ask a photographer what the biggest limitation of today's camera is, he will probably answer
without any hesitation: dynamic range. So, because of this limitation, a photo of any high or medium
dynamic range scene will end up with blown out highlights or undistinguishable shadows or both.
Therefore, an image that does not show this limitation, i.e. that reproduces the details in highlights
and shadows of the original HDR scene the same way a human observer had seen it, such image
does not look like what people expect a photograph to look like. In this sense, such image is not
photorealistic.

Photographers are not happy with the limitation of their camera, and this is what I meant by
"photographers do not wish to produce photorealistic images". Photographers would like that their
camera were able to reproduce that beautiful HDR landscape that captivated their eyes, but they
know the camera will not collaborate...

This is why the promises of HDR photography are so exciting for photographers. They are all looking
forward for a camera that will give them directly an HDR picture. (Please note that I am intentionally
using the term "picture" instead of "image" here, to make it clear that what photographers are
interested in is the final tone mapped output and not the "raw" 32-bit HDR image itself.)

By saying "HDR picture" a photographer would thus mean a tone mapped HDR image. Such image is
technically speaking an LDR image, but for a photographer it reproduces the high dynamic range
captured, and is thus referred to as HDR. A photographer will consider such tone mapped image as a
good "HDR picture" if it succeeds in reproducing the highlights and shadows as he/she saw them at
capture time. If the tone mapped image looks like a "photorealistic" blown-out image instead, it
won't be considered "HDR" for a photographer.

Let's take this example to illustrate that
http://www.hdrsoft.com/images/grandcanal/tm.html

Most people looking at the above tone mapped image (especially when they are not photographers
themselves) will typically react this way:
"This picture does not look natural. It looks more like a painting than a photograph".

However, if you tell that to Jacques Joffre, the photographer who took the bracketed shots for this
picture, he will reply:
"But I was there and this is exactly what I saw! It is precisely because it looked so beautiful that I
decided to shoot this scene. I would not have done it if it had looked like a dull photograph."

I won't claim that the tone mapper of Photomatix always succeeds in closely reproducing what the
photographer saw. This depends on the image and the dynamic range, sometimes it works very well,
sometimes it does not, and other tone mapping operators -or simply blending the original
exposures- will work better. And of course, a user can always intentionally choose settings that
make the image look "special" and far away from what it was in reality. But the same can be done
with any TMO that offers sufficient user control, it is unfair to blame only Photomatix for that.

My point is just that, in my opinion, we should not rate a tone mapping algorithm on how
photorealistic its output is perceived to be, but rather on how well it reproduces what a human
observer has seen.

Geraldine Joffre

···

On Fri, 11 Aug 2006 10:42:38 -0400, Kirk Thibault wrote

On Aug 11, 2006, at 4:36 AM, Geraldine Joffre wrote:
<snip>
>
> Photographers in general do not wish to produce photorealistic images.
> Geraldine Joffre
>
</snip>

This is a very interesting statement, especially in light of 3D
modelers and renderers who are trying to produce "photorealistic"
images by using HDR photographs as radiance maps and reflectance
maps. Fascinating philosophical point of view! I guess it comes
down to one's definition of what is "photorealistic".