Luminous Landscape Forum

Raw & Post Processing, Printing => Digital Image Processing => Topic started by: paulbk on April 05, 2006, 08:22:02 pm

Title: DCF Full Spectrum plug-in
Post by: paulbk on April 05, 2006, 08:22:02 pm
Has anyone tried this?  DCF Full Spectrum (http://www.tribecalabs.com/examples_video_lg.htm)

This from Jon Cone's Inkjet Mall (http://shopping.netsuite.com/s.nl/c.362672/sc.14/category.529/.f)... "This unique technology allows your camera to reproduce all colors accurately, including difficult hues such as violet, deep blue, and sunlit green.

The RGB (Red, Green, Blue) color model was designed for the efficient production of colors using a trichromatic light emitter, such as a computer monitor or a camera LCD. It is a model for producing illumination; it was not designed to simulate the complexity of human color vision. Although RGB can be used to produce millions of color combinations, the hues are not arranged in a way that is consistent with how we see. As a result, digital cameras generate a simplified, limited spectrum."
Title: DCF Full Spectrum plug-in
Post by: gryffyn on April 07, 2006, 12:36:00 pm
Quote
Has anyone tried this?  DCF Full Spectrum (http://www.tribecalabs.com/examples_video_lg.htm)

I have it.  It was inexpensive so I thought it would be a good addition to the CS2  "arsenal", but I haven't used it enough to really comment whether it is worth the effort or not.
Title: DCF Full Spectrum plug-in
Post by: Digi-T on April 07, 2006, 03:37:26 pm
This looks interesting but I don't want to spend even 50 bucks on a product where I can't test it first on my own images. There needs to be a trial version of this product. I couldn't find any sort of reviews on it either.

T
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 07, 2006, 03:57:58 pm
Quote
Has anyone tried this?  DCF Full Spectrum (http://www.tribecalabs.com/examples_video_lg.htm)

This from Jon Cone's Inkjet Mall (http://shopping.netsuite.com/s.nl/c.362672/sc.14/category.529/.f)... "This unique technology allows your camera to reproduce all colors accurately, including difficult hues such as violet, deep blue, and sunlit green.
[{POST_SNAPBACK}][/a] (http://index.php?act=findpost&pid=61967\")

What has Jon been smoking <g>

This was discussed in length on the ColorSync list (see URL below). The color geeks dismissed it as flat out BS! Don't waste your money or your time.

[a href=\"http://lists.apple.com/archives/colorsync-users/2006/Mar/msg00245.html]http://lists.apple.com/archives/colorsync-...r/msg00245.html[/url]
Especially useful:

http://lists.apple.com/archives/colorsync-...r/msg00249.html (http://lists.apple.com/archives/colorsync-users/2006/Mar/msg00249.html)
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 07, 2006, 04:36:38 pm
Every Macbeth Color Checker I've ever seen has a nice purple patch (Patch 10 and not too far off, patch 17). So I took my wife's Digital ELF PHD camera and shot one. This isn't what I'd call a high end camera guys. Do you see anything wrong with those patches?
Title: DCF Full Spectrum plug-in
Post by: paulbk on April 08, 2006, 01:13:22 pm
re: DCF...... Save your money!

I start with a simple extracted JPG from 1D mk2 RAW file, same file processed in both cases, and ONLY the corrections noted:
LEFT chart... DCF Full Spectrum Correction applied.
RIGHT chart... only levels with snap neutral midtones applied.

Looks like DCF turns blue to purple. Pathetic.
Title: DCF Full Spectrum plug-in
Post by: Graeme Nattress on April 08, 2006, 08:09:48 pm
When you go from RAW from the sensor, to RGB in Photoshop, there are a few things that happen along the way. One is the colour matrix, which is a 3x3 set of values that transforms the colour space of the camera with respect to the incoming light and the RGB filters on the pixels into the RGB space of whatever you choose your RGB space to be. Looks to me that DCF are tweaking those values to use a different RGB transform matrix. Look slike nothing you can't do with Photoshop and a bit of time on your hands to get the exact values, then make it an action....

Graeme
Title: DCF Full Spectrum plug-in
Post by: 61Dynamic on April 08, 2006, 11:50:41 pm
It's obvious just by looking at their sample images that the program is doing nothing more than shift the hue of blue. But hey, it has contrast and hue adjustment sliders too so it must be worth $50.
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 10, 2006, 07:23:24 am
Mr. Rodney, thank you for providing such a good image of the Macbeth Color Checker so that I can demonstrate the effects of DCF Full Spectrum.

DCF Full Spectrum is a nonlinear RGB color palette for designed to simulate the complexity of human color perception.

The RGB model used by all digital cameras is a model designed for the efficient production of illumination using three colored stimuli. Colors are spatially and perceptually even fractions of the whole. They remain chromatically constant as their value is changed. Human response to a color, on the other hand, is not usually described as spatially and perceptually even fractions of the response to illumination. We have a much more complex processor than that. We know intuitively that as a color becomes darker it becomes deeper. This is a human characterization. A digital camera does not characterize, it can only quantify. The result is that darker colors, like purplish blues lack the characterization of purplish and simply get displayed as dark blue. (If you have ever watched the LA Lakers or the Minnesota Vikings on TV, you will know what I mean by the above. On TV, their jerseys appear dark blue, but in real life their jerseys are deep purple.)

DCF Full Spectrum characterizes the data quantified by a digital camera to produce more photorealistic images.

The image on the left is the image Andrew Rodney provided, the image on the right uses DCF Full Spectrum. As pointed out, the most obvious difference is that the DCF image has deeper blues (or to use their descriptive ISCC/NBS names “purplish blues”). Additionally you may also notice warmer, deeper greens and oranges. In fact, closer examination reveals that all colors go through a transformation. Improved color contrast, the perceived brightness of adjacent colors, means DCF images looker sharper and more clear with no sharpening.

While we are in development of a water mark version of DCF Full Spectrum, we will happily process test images for those who wish to evaluate results before making a purchase, for information e-mail contact@tribecalabs.com.

Thank you,

Mike Bevans
Tribeca Labs
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 10, 2006, 11:21:00 am
I'm looking at both files after downloading and assigning sRGB (not the silly camera RGB profile the toy Canon tagged, should have done that first) and in Photoshop, I'm not seeing anything the image to the right really brings to the party. Certainly when comparing it to a LAB Macbeth version and certainly not withstanding a tweak (if anyone even thinks it's necessary) with selective color.

The web site makes it pretty clear this some heinous issue with digital cameras and this $50 product is a godsend but I still think, based on this test that you've got a solution in search of a problem.

Least we forget, this is a Macbeth JPEG from a $200 Digital Elf circa 2003 or so.
Title: DCF Full Spectrum plug-in
Post by: Loadus on April 10, 2006, 11:25:44 am
Curious reading about that hue error compensation that the plugin does. Sounds like it's trying to repair the gamma error damage from 2.2 boost. If you work on linear intensity space this plugin probably is not worthwhile. (I probably should test it myself before commenting ...)
Title: DCF Full Spectrum plug-in
Post by: paulbk on April 10, 2006, 08:35:50 pm
Mike Bevans, Tribeca Labs

See my examples below. Why does DCF Full Spectrum correction turn blue (col 1, row 3) to purple-ish?

I've tried this many times using different RAW convert software and making no corrections in PSCS2 other than DCF. My color checker was shot in bright sun using Canon 1D Mark II. I generally like what it does to all other colors except the shift in blue to purple. Can't use it with a clear blue sky, it makes it purple-ish.

RAW.. white balance, convert to color space =  Adobe1998
LEFT target: Photoshop.. no correction
RIGHT target: Photoshop.. DCF Full Spectrum applied
Both files converted to sRGB for web JPG examples.
sRGB looks no different then Adobe1998 on my calibrated monitor.

p
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 14, 2006, 07:58:45 am
Paul,

DCF Full Spectrum is a color palette designed for digital photography. Just like back in the day when you would choose between Kodak and Fuji film, you were choosing different palettes. Both Kodak and Fuji film could be properly exposed and balanced; however, it was commonly believed that Kodak had deeper blues and Fuji had more vibrant reds. The color palettes between the two films were different.

DCF Full Spectrum is an alternate RGB palette for digital photography. DCF Full Spectrum is our “Reala” film, DCF Portrait is our “Portra” and DCF Vivid is like “Velvia”. Using the DCF Control Panel, you can customize your own color palette.

Before I begin an analysis of your results, a couple things should be pointed out. You mention that you shot the color target in bright sunlight. If this is the case, your starting exposure is a little bit down. According to the folks at gretag, the white square on the color chart will read about 243RGB when properly exposed, whereas the white patch on your color target reads about 220RGB. Also, it should be noted that even though your white balance is correct the appearance of the blue squares in question will vary when removed from sunlight and viewed inside next to a monitor. White may still look white (because the brightest white in the scene is your eye’s white point) but the particular blue square you mention will appear much redder when in sunlight (depending on the tiome of day) than it does indoors. I mention this simply to point out that we don’t have an apples-to-apples test to begin with.

That said, you are correct, DCF Full Spectrum blues are more purplish. The blue you refer to is actually called "vivid purplish blue" (and the blue patch up and to the right from this one is referred to as "strong purplish blue") by the Inter-Society Color Council (ISCC) and the United States Department of Commerce's National Bureau of Standards (NBS) (now called the National Institute of Standards and Technology).

DCF Full Spectrum is a color difference model between RGB and human color perception. It is our definition of digital color fidelity. How much your particular camera differs from our default setting may vary, based upon camera performance(exposure, for example) and individual preference. If you feel the color is too purple, you can use the DCF Control Panel to define your own settings.

By turning OFF DCF Full Spectrum in the Control Panel and turning ON the Additional Controls, you can set your own difference model using our color formula (to varying degrees). Set each of the four color sliders to 50% resulting in a formula equal to our default DCF Full Spectrum. I recommend setting all of the sliders at their 50% position (equal to our default) but turning down the Blue slider to 33-25%. You will still see deeper blues and purples, but not quite as strong as our default.

Once you have a setting you like, you can create a Photoshop Action that LOADs your preferred setting. (To record this action, open a dummy image, click the LOAD button in the DCF Control Panel and load your preferred setting). Once you have done this, you can simply run the action without needing to go through the Control Panel step.

Thank you,

Mike Bevans
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: marc.s on April 14, 2006, 08:44:22 am
Quote
DCF Full Spectrum is an alternate RGB palette for digital photography. DCF Full Spectrum is our “Reala” film, DCF Portrait is our “Portra” and DCF Vivid is like “Velvia”. Using the DCF Control Panel, you can customize your own color palette.
[a href=\"index.php?act=findpost&pid=62546\"][{POST_SNAPBACK}][/a]

There's a difference between a product that adds a number of styles - or film emulations - and then claiming that digital cameras do not record purple, is there not?

Despite the claim on the front page of your website: "Your eyes see purple. So why doesn't your camera" both my digital cameras are perfectly capable of recording purple.

There's nothing wrong with your product, only with your fraudulent advertising.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 14, 2006, 10:10:16 am
Quote
There's a difference between a product that adds a number of styles - or film emulations - and then claiming that digital cameras do not record purple, is there not?

Despite the claim on the front page of your website: "Your eyes see purple. So why doesn't your camera" both my digital cameras are perfectly capable of recording purple.

There's nothing wrong with your product, only with your fraudulent advertising.
[a href=\"index.php?act=findpost&pid=62551\"][{POST_SNAPBACK}][/a]

I think you hit the hammer on the nail here! Good job.

I'm shooting RAW, and I have all kinds of tools at my disposal for producing a myriad of renderings (flim looks plus!).  Check out the HLS controls in Lightroom. Talk about a great tool for tweaking all kinds of renderings (and yes, you can do this on rendered files as well).
Title: DCF Full Spectrum plug-in
Post by: marc.s on April 14, 2006, 10:49:54 am
Oh, I'm so looking forward to Lightroom being available on Windows! Right now I have to tweak all my important pictures individually in Photoshop since I'm not happy with the controls of any current raw converter. The resulting file sizes (and time spent) are less than ideal, and from what I've heard Lightroom sounds like a big step in the right direction.
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 16, 2006, 05:10:31 pm
Thank you for the constructive criticism regarding our website. We will certainly take your comments into consideration as we redesign our website and fine-tune our message.

As we are all aware, color is a complex subject, and reducing the discussion to a simple statement appears to have caused some controversy. In no way do we intend to deceive, we simply want to make a case for a real solution to a real problem.

The problem addressed by DCF Full Spectrum is that the RGB system reproduces an incomplete spectrum of colors. As you can see in the digital recording and reproduction of the component colors of sunlight (the spectrum shot below), RGB systems have difficulty reproducing shorter wavelengths of light (original image on the left). In the Full Spectrum RGB image on the right, you can clearly see how the spectrum includes the shorter wavelengths of light for a more complete reproduction of violet, and in turn the whole spectrum.

[attachment=448:attachment]

This is what we mean when we say: “Your eyes see purple, so why can’t your camera?”

The poor sensitivity to violet can also be seen in this image of the LA Lakers jersey. If you have ever seen the Lakers on television, you will see that there jerseys appear bright blue, like the original digital capture on the left. The Lakers jerseys are purple, like the Full Spectrum RGB image on the right.

[attachment=449:attachment]

We first noticed the shortcomings of RGB in digital fine-art reproduction where color matching is critical. Tribeca Labs’ color experts have worked extensively with museums and cultural institutions involved in large scale digital preservation projects since 1998. Working in fully color managed environments with the highest resolution digital scan back cameras, we have been able to confirm that the problem is common and can be seen in every camera or monitor.

For nature photographers, DCF Full Spectrum provides richer, more photorealistic and natural colors, as you can see in the following images of flowers (original on the left, Full Spectrum RGB on the right).

[attachment=451:attachment]

[attachment=452:attachment]

Mark.S asks, “There's a difference between a product that adds a number of styles - or film emulations - and then claiming that digital cameras do not record purple, is there not?”

The comparison between Kodak film (deep blues) and Fuji film (vibrant red) was made to illustrate the concept of a color palette. For example, to say Kodak film has deeper blues than Fuji film is like saying Full Spectrum RGB has deeper blues than RGB alone.

Admittedly, one problem we have with our advertising is that people think that DCF Full Spectrum only affects purples. In the shot of the trees below you can see how DCF Full Spectrum brings out the complexity of greens, producing a better sense of space and dimension (RGB on the left, Full Spectrum RGB on the right).

[attachment=453:attachment]

Full Spectrum RGB is available in two additional settings, the “number of styles,” to which Mark.S refers: DCF Vivid and DCF Portrait. These are optional intensity settings that make use of the expanded Full Spectrum RGB color palette. The following photographs make use of DCF Vivid color (original image on the left, DCF Vivid on the right).

[attachment=454:attachment]

[attachment=455:attachment]

[attachment=456:attachment]

The above images are compliments of the photographers at www.sxc.hu.

Finally, because we know photographers like to make their own adjustments, we added the DCF Control Panel that allows the user to compose his own color palette and intensity settings.

I hope this clears up any misunderstandings. If you have any questions, you can contact me directly at mike@tribecalabs.com.

Sincerely,

Michael Bevans
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 16, 2006, 05:47:29 pm
Still don't get it. Notice my simple Photoshop fix using selective color. Two sample points in Info palette with before and after numbers show my simple tweak produced the same fix. This "fix" assumes I actually prefer the color on the right side compared to the left in the original file. Bottom line however, it's super fast and simple to produce this rendering in Photoshop.
Title: DCF Full Spectrum plug-in
Post by: Graeme Nattress on April 16, 2006, 05:57:23 pm
Digitaldog, you should turn that into an action and sell it! :-)

Graeme
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 16, 2006, 06:11:44 pm
Mr. Rodney,

Thank you for pointing out that there are many ways to skin a cat.

Just as I can purchase a Photoshop plug-in that gives me a cross-processed look, I can just as easily get the same result with a number of Photoshop tools, like Channel Mixer and/or curves. Similarly, I can purchase a plug-in that gives me a better neutral balance, an infra-red look, or a "nocturnal look", or I can spend my time in Photoshop using the tools available to get the look I want.

Obviously there are any number of tools that allow retouchers to manually adjust the colors of individual images.

The point of any program is to make a process easier. If you prefer to spend your time editing each individual picture, you can,or you can purchase a program that saves you time and effort and may well produce a superior result to manual intervention.

Happy Easter,

Michael Bevans
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 16, 2006, 06:56:34 pm
Quote
Just as I can purchase a Photoshop plug-in that gives me a cross-processed look, I can just as easily get the same result with a number of Photoshop tools, like Channel Mixer and/or curves.

One is infinitely easier to do than the other even for novice users. And while I agree that there are plenty of ways to produce effects in Photoshop (using its tools or others), its all about price to performance ratios. Again, I'm seeing what appears to be a solution in search of a problem.
Title: DCF Full Spectrum plug-in
Post by: 61Dynamic on April 16, 2006, 07:12:05 pm
Your program is not what you are claiming. You state there is an issue with how cameras record purple, but there is not as demonstrated by the photos of the Color Charts. The difference between what people mistakenly see as purple and what the camera records is an optical illusion (two separate colors - blue & magenta - blending together).

Your program fixes that "issue" by implementing a hue change. Something you admit can be done in PS. Granted there is nothing wrong with offering another means of doing so, you are claiming there is this issue when the camera is actually recording purple. In addition you are making vague (and disingenuous) claims of using a larger color spectrum.

So my question is this, which is it really? Is it a matter of just adjusting hue, or are you doing some fancy conversions using color profiles of some sort? If the latter, lets here some technical explanations. What are the RGB cords, gamma and temp of the color space you work with, what is the frequency range in numbers of the "full spectrum" you are interpolating into the image, and how are you implementing this supposed spectrum?

You claim cameras don't use a full spectrum in the fact they don't record the shorter wavelengths of light. How could you use a full spectrum in your corrections when the camera didn't record the information to begin with? For us to believe your primary claim, we must first have to accept that your software is able to create something from nothing. Not only that, but that something is a color of light that you couldn't even possibly know existed to begin with.

All of the other adjustments made by the program are simple hue, saturation and contrast adjustments.

The summary here is that you are claiming to use this mysterious spectrum of color in some vague way to correct a problem that does not truly exist. Lets hear some substantiated information to back up your claims rather than vague references and unidentified "experts."
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 16, 2006, 07:14:46 pm
Quote
The summary here is that you are claiming to use this mysterious spectrum of color in some vague way to correct a problem that does not truly exist.[a href=\"index.php?act=findpost&pid=62747\"][{POST_SNAPBACK}][/a]

I've asked the always unflappable Dr. Karl Lang () to comment on all this "color science". I hope he pings us about this so called "issue" with digital cameras and "RGB". That will make this post quite entertaining (and educational).
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 16, 2006, 08:48:47 pm
Happy Easter.

Michael Bevans
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: dwm1953 on April 16, 2006, 10:21:14 pm
Mike, I applaud your patience in explaining your product to a skeptical audience. I have seen other company reps lose their cool under similar circumstances. I just spent some time at your website looking over your software. I see that the plug-in works within RGB and re-maps values in the uncorrected image to new values for the "correct" image produced by Full Spectrum. Nothing magical. Something an experienced photoshopper could do given the time and inclination. Just as many other plug-ins, it automates a process that can be done manually.

I think there is something to your argument that purples are not well reproduced (I have experienced this myself), however whether this is due to an inherent limitation in the RGB model or not, is not clearly demonstrated at your site. You provide anecdotal justification for the hypothesis but no real scientific explanation. This leads to skepticism and an unwillingness to part with $50 without first giving the plug-in a spin. As someone else pointed out (and you alluded to) a demo version is definitely needed here.
Title: DCF Full Spectrum plug-in
Post by: paulbk on April 16, 2006, 11:04:19 pm
re: DCF

Michael Bevans,
The more I play with DCF the more I like it. You need better documentation. You need to tell more sophisticated users "what exactly" each control is doing. For example, is the blue slider making the "b" curve in Lab steeper? Give me some indication what I'm really doing with out giving away the store.

Anyway, good luck with DCF. There may be a user base out there but you gotta upgrade the documentation.
paul
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 17, 2006, 08:50:18 am
Quote
I think there is something to your argument that purples are not well reproduced (I have experienced this myself), however whether this is due to an inherent limitation in the RGB model or not, is not clearly demonstrated at your site.

I've seen a fair number of output profiles shift blues to magenta or cyan but this is more an issue of a profile not really doing a very good job of handling either gamut mapping or handling blues. Going back say 5-6 years ago, there were few profile packages that handled this blue issue well. Today it's the exception to the rule.
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 18, 2006, 02:54:46 pm
Once again, great questions and comments. I will try to address each in turn. I am sure that my answers will raise more questions and comments.

Thank you, dwm1953. It is true that new technology is often greeted with skepticism at first. I recall in 2000, when I was a technical-rep for a digital camera-back manufacturer, introducing digital cameras to reluctant professional commercial photographers. Now digital cameras outsell film cameras. Opinions change quickly. I appreciate this forum as an opportunity to share information.

“You provide anecdotal justification for the hypothesis but no real scientific explanation.” The scientific explanation is simple; we observed a well documented problem, devised a theory to solve the problem and tested (and tested and tested). In short, DCF Full Spectrum is our proprietary color specification model designed to overcome color deficiencies identified by means of a full-blown psychophysical analysis of color difference between two color systems, RGB and human color perception. It is a color appearance model, or what I have earlier referred to as a palette of colors. To address the concerns of 61 Dynamic, DCF Full Spectrum is entirely device independent and will work with any color space. It does not alter balance, exposure or color settings.

I first noticed a deficiency in digital capture in 1998 at the Herbert F. Johnson Museum of Art at Cornell University working on one of the first large-scale digitization and preservation projects in the country. Over the course of two years, along with two other photographers, cataloguers, technicians, administrators, and museum staff, we set about the ambitious task of digitizing the entire collection of the museum, over 25,000 works of art. Scanning and quality checking over 65 high resolution images a day in a fully color managed environment, I began to notice that digital capture had a difficult time reproducing certain hues, regardless of the attention spent on proper balance and exposure, and regardless of the calibrated monitors and daylight balanced full spectrum lighting. It was curious, because I could create all the colors on the computer (I remember plotting colored points using Basic in junior high school), but the camera/monitor system could not properly characterize wram greens, deep blues, indigos and of course, purple.

Robin Myers, inventor of ColorSync confirms this observation in a very good article about the digital reproduction of art in which he notices a problem with reproducing cobalt blue, http://www.betterlight.com/pdf/whitePaper/...urate_photo.pdf (http://www.betterlight.com/pdf/whitePaper/wp_color_accurate_photo.pdf). In this article, he addresses 61 Dynamic’s question regarding what the camera actually records. Mr. Myers states that “The basic camera sensor is panchromatic.” A sensor quantifies all wavelengths of light, using filters to limit the sensitivity of a sensor to certain portions of the spectrum. The sensor itself does not see colors; it quantifies volumes of light data. The RGB colors that you see on your screen are mapped using CIE colorimetry.

Charles Poytnon, Fellow of the Society of Motion Picture and Television Engineers and author of the ubiquitous ColorFAQ and GammaFaq explains, “Color in the real world is best described in terms of distribution of power across the spectrum of visible light. Human vision maps these power distributions into sensory values, then processes these signals at successively higher and higher levels. The famous CIE color matching functions define the mapping from spectral power distributions (SPDs) to tristimulus values; these values are then the basis of color systems used for measurement and image coding. However, psychophysical data has now revealed the "cone fundamentals" that are taken to be the raw spectral sensitivities of human vision. The cone fundamentals don't quite match the CIE color matching functions,” or more simply, “Existing, practical cameras and scanners have spectral responses that don't closely resemble either the cone fundamentals or transformations of the CIE color matching functions. Therefore, these cameras see some colors differently from the way that vision sees those colors.” - http://www.poynton.com/notes/bio/goals.html (http://www.poynton.com/notes/bio/goals.html)

DCF Full Spectrum does not create something out of nothing; it remaps data quantified by the sensor. As dmw 1953 states, this is nothing magical; it just hadn’t been done before. Maybe my explanation has provided insight as to why. Certainly if you want to manually color match individual colors using an image editor, there are a million ways to do this. DCF Full Spectrum is a comprehensive automatic solution to an industry wide problem so you don’t have to.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 18, 2006, 03:32:49 pm
Quote
DCF Full Spectrum does not create something out of nothing; it remaps data quantified by the sensor.

The sensor is simply a photon counter and is color blind (RAW data is essentially Grayscale data).

How is the data quantified given this fact?
Title: DCF Full Spectrum plug-in
Post by: 61Dynamic on April 18, 2006, 06:53:14 pm
Quote
The sensor is simply a photon counter and is color blind (RAW data is essentially Grayscale data).

How is the data quantified given this fact?
[a href=\"index.php?act=findpost&pid=62981\"][{POST_SNAPBACK}][/a]
Given that information and the fact that each camera has its own interpretation of color due to variations in the color filters used on the sensor and the variables in the numerous means to convert the raw data to a working image file (each with it's own interpretation of the original luminosity data), how can DCF-FS possibly begin to know how to properly correct the color the camera recorded?

How can it know what needs to be "quantified" and how it needs to be "quantified" when it hasn't a clue what the original color could have been?

Does your software have specific color adjustment profiles for each camera made using each of the possible means of raw conversion in order to know how the adjustments should be made? Or does it just assume that all blue hues it receives should have more magenta in them regardless of what it is processing?
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on April 19, 2006, 06:54:01 pm
It is true, a digital camera is only a photon counter. Digital cameras do not see colors. As the Robin Myers article details, “The basic digital camera sensor is a panchromatic sensor; it responds to light from the ultraviolet through the visible, and well into the infrared spectrum. To produce a color sensor, filters are applied to its surface to give the sensor color selectivity. Each filter limits the panchromatic response of the silicon sensor to a small portion of the spectrum.” http://www.betterlight.com/pdf/whitePaper/...urate_photo.pdf (http://www.betterlight.com/pdf/whitePaper/wp_color_accurate_photo.pdf)

Charles Poynton’s Guide Tour of Color Space provides more information on transforming input voltages into color signals in his article “A Guided Tour of Color Space.”
http://www-scf.usc.edu/~csci576b/Slides/Le...or%20space' (http://www-scf.usc.edu/~csci576b/Slides/Lecture4_ColorSpaces.pdf#search='tour%20of%20color%20space')

And, finally, for an excellent interactive on how to build a CCD you can check out this page: http://micro.magnet.fsu.edu/primer/digital...ccdanatomy.html (http://micro.magnet.fsu.edu/primer/digitalimaging/concepts/ccdanatomy.html)

In any event, suffice it to say that input voltages are converted to color values using a color specification system based on CIE colorimetry (otherwise we’d only have black and white digital cameras).

As already described by Poynton in an earlier post; however, this system is necessary for encoding digital images but not sufficient to produce colors that accurately simulate human color perception.

To answer Dan’s question, DCF Full Spectrum is a proprietary device independent color model. It does not rely on data about the camera system in order to transform colors. For more information on how DCF Full Spectrum characterizes colors, refer to http://www.tribecalabs.com/technology.htm (http://www.tribecalabs.com/technology.htm).

-Mike Bevans
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 19, 2006, 07:17:43 pm
You're doing a fine job of providing links to useful articles that in no way backs up ANYTHING you've said about your product. This is getting silly too. You've got RGB data and all you can do is change the numbers one way (and affect other numbers too). This isn't anything at all different from my Selective Color tweak in Photoshop where I took some existing RGB numbers and altered them in one direction or the other. All this talk about proprietary this and that doesn't really change the fact that you're not doing anything at all unique. Now if you want to tell folks "we have a easy button that shifts blues one way or the other in the event you find alternative controls in Photoshop diffuclt" I'm OK with that. But there's a great deal of smoke and mirrors and a huge BS factor on your web site with very little to back up anything other than "we have a make pretty button for blues."

If your idea of selling this "technology" is to have someone photograph your web page and view the results on their cameras LCD, I think you're talking to the wrong crowed here.

So this is some proprietary device independent color model is it? So you're doing color space conversions? Or, as I suspect you're simply altering selective color numbers in one direction which really isn't all that unique (Photoshop has been doing this since 1989 before it was called Photoshop.
Title: DCF Full Spectrum plug-in
Post by: 61Dynamic on April 19, 2006, 09:39:06 pm
Quote
To answer Dan’s question, DCF Full Spectrum is a proprietary device independent color model. It does not rely on data about the camera system in order to transform colors. For more information on how DCF Full Spectrum characterizes colors, refer to http://www.tribecalabs.com/technology.htm (http://www.tribecalabs.com/technology.htm).
[a href=\"index.php?act=findpost&pid=63111\"][{POST_SNAPBACK}][/a]
As I suspected. DCF-FS is simply a simplified Hue/Saturation/Contrast tool. Perhaps it's using proprietary variation of CIE Lab to do so, but it certainly is not a tool that is capable of correcting this "problem" you keep describing.

As Rodney pointed out, you are throwing out a fat load of BS to make it sound like your program is allot more than it really is.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 20, 2006, 09:16:33 am
Quote
As Rodney pointed out, you are throwing out a fat load of BS to make it sound like your program is allot more than it really is.
[a href=\"index.php?act=findpost&pid=63124\"][{POST_SNAPBACK}][/a]

I feel a bit guilty that I'm busting Mike's balls (well I am a bit). Again, if the intent is to make it easy for novice users to "fix" an issue with blues (which I submit has nothing to do with digital sensors per say and have shown with a $200 point and shoot), fine. But the message is getting lost in the high signal to noise ratio which at this point does appear to follow your above description (a fat load...).

For the life or me, I can't find anything my quick and dirty tweak in Photoshop didn’t accomplish with respect to fixing the blues. But I fully admit that some users may not know how to do this (although it's pretty darn simple and a one page tutorial would set anyone willing in the right direction).
Title: DCF Full Spectrum plug-in
Post by: Graeme Nattress on April 20, 2006, 11:29:20 am
Adjusting Hue in Photoshop with the Hue and Saturation adjustment layer is easy. Just set it to blues only, slide the hue a bit to add some purple into the blues and you're done. Adjust to taste - yummy. Quick, easy and painless. You could even make it an action if you've got preferred settings.

I'd totally agree that this is  not a camera issue. The gamut of colours that a camera sees is just fine, but yes, there is no one perfect colour matrix to convert that image the camera sees to full RGB colour. That's where your photography eye comes in, to adjust your image in post to your liking.

What I want to know about DCF, is if you take a picture of something blue, and it looks blue in Photoshop, then applying the DCF will turn it, or part of it, purple, and that would be less accurate. How does it tell blue from blue that stays blue, and blue that's really purple. And given that each camera sensor's colorimetry is different, and each gets turned to RGB via a different colour matrix, how can it ever be accurate? If it is, as I surmise, purely perceptual and subjective, then tweaking the hue to taste, as I outline above would be the easy and free solution.

Graeme
Title: DCF Full Spectrum plug-in
Post by: paulbk on April 20, 2006, 05:21:37 pm
DCF example test files below:
Camera: Canon 1D Mark II, RAW mode
RAW Conversion: default Adobe Camera RAW auto-mode, 16 bit, Adobe(1998)

Converted RAW file was duplicated.
Left file -- no further processing after RAW conversion.
Right file -- only "default" DCF Full Spectrum processing applied after RAW conversion.
DCF adjustments are available in "Control Panel" mode. I didn't use them. If I'm going to tweak, I'll do it myself using a Lab curve without use of DCF.
Each file resized to jpg, sRGB, and posted.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 20, 2006, 05:40:37 pm
The sky on the left looks better to me. But either would fly and the differences are not much. It looks like the nice blue sky in the left went magenta in the right...
Title: DCF Full Spectrum plug-in
Post by: Graeme Nattress on April 20, 2006, 07:02:19 pm
I think the image that has not had DCF applied looks better. The sky is purple in the DCF one (see my comment about it cannot know which blues are really blue and which are really purple) and the green grass has turned brown.

Graeme
Title: DCF Full Spectrum plug-in
Post by: 61Dynamic on April 20, 2006, 08:27:35 pm
Quote
I feel a bit guilty that I'm busting Mike's balls (well I am a bit). Again, if the intent is to make it easy for novice users to "fix" an issue with blues (which I submit has nothing to do with digital sensors per say and have shown with a $200 point and shoot), fine. But the message is getting lost in the high signal to noise ratio which at this point does appear to follow your above description (a fat load...).
[a href=\"index.php?act=findpost&pid=63164\"][{POST_SNAPBACK}][/a]
I don't feel guilty. If Mike was making these claims out of honest ignorance, then that would be forgivable. I have a hunch that is not the case since the entire company seems built upon this one fallacy. He makes certain claims and it is up to him to prove them yet the guy continually throws out the same uninformative or off-subject garbage in his replies.

If a company is not going to be upfront and honest, then by all means they deserve every ounce of criticism they get. I have little tolerance for any business, big or small, that tries to take advantage of people who don't know any better.

Maybe he'll prove me wrong with his next reply(s) if he decides to continue in this thread.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 20, 2006, 08:29:31 pm
Quote
... the green grass has turned brown.
[a href=\"index.php?act=findpost&pid=63202\"][{POST_SNAPBACK}][/a]


Good call! I was too quick to view the two images and missed the grass. That's absolutely a deal breaker in the 2nd image. So I think you're spot on; you alter the image that doesn't need alteration, you can hose the file. BTW, a phone conversation I had with Karl Lang yesterday about this topic confirmed what you're saying: you have to move the values in one direction or another and assuming the image is OK to begin with, such a tool can produce as many problems as it hopes to solve.
Title: DCF Full Spectrum plug-in
Post by: nemophoto on April 24, 2006, 12:01:57 pm
Frankly, all this is "much ado about nothing". Either use the pluggin. . . or not. It does work. I shoot fashion where color is always a problem with garments. Most files are handed to clients in AdobeRGB (their request). In my view, I've done my job when I can get the garments as close as possible. It's the pre-press house that needs to do final color matching.

Andrew's example of how he matched the purple-ish Laker's jersey has a flaw. Sure, he matched the purple, but the gold was still off and greenish on my monitor (yes, calibrated).

The point is, you can chose to buy the plugin or not. I've bought lots of pluggins and actions over the years that have proved either ineffective or a waste of money. Case in point: Andrew's own PixelGenius enterprise produces PhotoKit Sharpener. I bought it. Used it some and decided I prefer using Nik Sharpener 2.0. It's not that PK Sharpener is no good, only my preference.

It's sad to see people be so bombastic in their approval or disapproval of products. There are lots out there that ARE total garbage. In this case, we're talking subtleties, not a "blow you away" pluggin. To me, a lot of this smacks of a "turf war" and who's the expert or not.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 24, 2006, 12:12:28 pm
Quote
Andrew's example of how he matched the purple-ish Laker's jersey has a flaw. Sure, he matched the purple, but the gold was still off and greenish on my monitor (yes, calibrated).
[a href=\"index.php?act=findpost&pid=63567\"][{POST_SNAPBACK}][/a]

Looks just fine to me (yup, calibrated display). Are the two sets of RGB numbers identical? Probably not. But that's a moot point. I could easily have messed with the yellows too. The color appearance after my edit doesn't seem to be an issue.

What IS an issue is the BS factor of the site selling this product. I think we've all been pretty clear that a plug-in that provides  a "make blues pretty" button is fine as long as you tell folks that's what it does. But to put up a bunch of bogus technical nonsense about how cameras can't "see" blue or suggesting you photography your display when viewing their site and examining the LCD on a PHD camera as evidence that there's some kind of color issue with digital cameras is insulting and well, plain BS.

No one is disputing the plug-in might have merit for some users who don't know Selective Color in Photoshop from the crop tool. If it fixes your blues and you find the $50 a worthwhile investment, far from anyone to fault that. But lets cut the Krap and stop providing URLs that don't back up the fact that the premise provided on the web page about this product is Krap.
Title: DCF Full Spectrum plug-in
Post by: PeterLange on April 24, 2006, 05:53:48 pm
Quote
“Existing, practical cameras and scanners have spectral responses that don't closely resemble either the cone fundamentals or transformations of the CIE color matching functions. Therefore, these cameras see some colors differently from the way that vision sees those colors.”


Nope; sorry I have to disagree.


Michael,

In the meantime state-of-the-art digital cameras were reported to have a metamerism index of above 90% or so (according to ISO 17321).  Since perfect color reproduction is scaled to a value of 100, any sensor that achieves a metamerism index above 90 should be capable of providing great color.

Actually the spectral response of Bayer R/G/B filters can come close enough to a linear combination of the CIE XYZ matching functions (depending on the camera of course). So that a simple 3 by 3 matrix can be used in software to recover the corresponding XYZ values, or for direct conversion to a common output space like sRGB, aRGB, pRGB ...  In other words, within a quite large 'triangular' matrix gamut digital cameras see nearly the same as the human obverver. Note, I don’t say perfectly.

Talking about: a preferred color reproduction, a pleasing rendition (output-referred), Color Appearance Models, etc. ... that’s a different story.

So where exactly lies the problem which your plug-in intends to cure?

Please be clear.

Peter

--
Title: DCF Full Spectrum plug-in
Post by: allan67 on April 24, 2006, 07:09:27 pm
I did a test today to see if there's really a problem with DSLR not seeing purple colours.
Used 30D to take a shot of my MTB club jersey - it has a very nice gradient from light blue through deeper blue to purple. Effectively, on the camera's LCD the colours really looked very much off - the purple became blue and blues were shifted to cyan. BUT, when the RAW file was opened in ACR 3.3 (calibrated for the shooting conditions using GMB chart as explained by Bruce and others) everything became very much normal - all hues snapped to their correct values and the screen to object match was nearly perfect (needed a bit of saturation increase to get exact match).
FWIW.
Allan
Title: DCF Full Spectrum plug-in
Post by: Hermie on April 25, 2006, 06:27:39 am
From the FAQ's on Munsell Color Science Laboratory:

Question:
"Digital image sensors (such as those used in digital cameras)use red, green, blue ink-based color filters to generate color. Do they therefore have a color gamut that limits the range of colors that they can detect? (255)"

Answer:
"Let's start with the short answer to your question; there is no such thing as a camera, or scanner, gamut. A gamut is defined as the range of colors that a given imaging device can display. To say that a camera had a gamut would be to imply that you could put a color in front of it that it could not possibly respond to. While it is certainly possible that two colors that are visually distinct might be mapped into the same color signals by a camera, that does not mean that the camera could not detect those colors. It just couldn't discriminate them. For example, a monochrome sensor will map all colors into a grayscale image and encode it as such. Certainly the encoding has a gamut (in this case a lightness range with no chroma information), but did the camera responded to all the colors put before it. It is the encoding that imposed the gamut. In the color world, encoding is based on some explicit or implied display. For example, sRGB is a description of a display and therefore defines a gamut (but only if the sRGB values are limited in range). If a camera encodes an image in sRGB, that doesn't mean that the range of colors the camera detected are only from within the sRGB display gamut, but it means the camera data have been transformed to best use that sRGB encoding. As long as a camera has three or more sensors that span the visual spectrum, then it will respond all the same stimuli as our visual system. Whether the camera can discriminate colors as well as the human visual system will depend on the encoding of the camera signals, quantitization, and the details of the camera responsivities. (To return to the black and white system, that camera encodes all the colors into a gray scale. They could then be displayed as any color within a given display, but many colors from the original scene would be mapped to the same values.)

Since there is no such thing as a gamut for an input device, then there is no way to compute it or calculate a figure of merit. Generally, the accuracy of color capture devices is assessed through the accuracy of the output values for known inputs in terms of color differences. Also, sensors are sometimes evaluate in terms of their ability to mimic human visual responses (and therefore be accurate) using quantities with names like colorimetric quality factor, that measure how close the camera responsivities are to linear transformations of the human color matching functions. Doing an internet search on "colorimetric quality factor" will lead you in the right direction."

Herman
Title: DCF Full Spectrum plug-in
Post by: jani on April 25, 2006, 07:47:49 am
Guys, the horse is nearly a wet puddle on the ground, it can't pull the plow anymore.
Title: DCF Full Spectrum plug-in
Post by: 61Dynamic on April 25, 2006, 11:39:32 am
Quote
Guys, the horse is nearly a wet puddle on the ground, it can't pull the plow anymore.
[a href=\"index.php?act=findpost&pid=63638\"][{POST_SNAPBACK}][/a]
No kidding, there's nothing left to beat.
Title: DCF Full Spectrum plug-in
Post by: PeterLange on April 25, 2006, 02:46:11 pm
Quote
From the FAQ's on Munsell Color Science Laboratory: ...

Hermie,

Please let me briefly comment on this Munsell quote, because I still believe that the whole story is much simpler than it seems:

The naked RGB triplets as released from Bayer interpolation *can* be assigned to a quite large matrix space as the input profile. To my best knowledge, that’s the way Camera Raw works. And this approach will work so much the better, the closer the spectral response of Bayer R/G/B filters comes to a linear transform of the CIE XYZ matching functions.

BUT – unlike output devices which have a rigid output gamut – this input gamut more describes a sphere of accuracy. Within this sphere the camera 'sees' about the same as the human observer. But, even outside from this sphere, a camera will probably respond to 'everything’ like e.g. spectrally pure laser beams. Just, this doesn’t lead to accurate data.

So if the Munsell lab likes to capture laser beams, then they’re right. There’s probably no camera and input gamut (which would be suited for such purposes). However, from a practical perspective the input gamut of today’s cameras is large enough for almost everything coming in front of my lenses.
 
 
Anyway, I’d still love to hear what Micheal Bevans from tribecalabs has to offer regarding the science part of his product. A nice ab initio explanation would be fine…

 Peter

--
P.S.: Did you notice my post on sensor design + reference literature before RG forums went down.
Title: DCF Full Spectrum plug-in
Post by: Hermie on April 25, 2006, 03:35:26 pm
Quote
P.S.: Did you notice my post on sensor design + reference literature before RG forums went down.
[a href=\"index.php?act=findpost&pid=63665\"][{POST_SNAPBACK}][/a]

No I didn't. Do you have any links for me?

Herman
Title: DCF Full Spectrum plug-in
Post by: Nick Rains on April 25, 2006, 05:49:15 pm
Equine flagellation aside...

It's interesting to note that Jon Cone sells the plugin on his site. I would imagine Jon can be assumed to know a thing or two about digital imaging.

I have the DCF plugin - Mike was kind enough to send me a copy for review and we exchanged emails about its effects.

I am no colour guru but I know what looks 'good'. My point to Mike was that I thought DCF made normal images 'less good' by making the blue skies too mag. and the greens too yellow. Now it may be that this is more 'accurate', but I personally don't want 'accurate' I want 'good' - not the same thing at all.

Blue skies with too much mag. are buggers to offset print, they are better to be slightly too cyan. However, digital camera do have a tendancy to exaggerate greens and Mike refers to the fluoro green effect. I like the bright greens but it is more accurate to tone them down and make them more yellow. Hence the example posted above is possibly more 'accurate' in the greens, but to my eye, I prefer the brighter green.

I was tempted to get out the GM chart but I thought what's the point? DCF does nothing for my files that I cannot do in PS and, on default, I don't like the effect.

Like Nemo says: get the plugin, check it out, if you don't like it, move on.

One effect that has not been mentioned is the Fill Light slider, a bit like RSP. This works very well with no mystery, and if some beginner had problems with certain tonal adjustments in PS then the plugin is probably worth having for this feature alone.
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 25, 2006, 06:37:42 pm
Quote
It's interesting to note that Jon Cone sells the plugin on his site. I would imagine Jon can be assumed to know a thing or two about digital imaging.

He knows a thing or two about selling... Other than that, I wouldn't assume anything. Or maybe Jon can come on down and take over for Mike about what this product is supposed to do (I submit again it's a make pretty button and there's nothing wrong with that. Let's just cut the technical nonsense off Mike's and Jon's site; both are used to sell the product and the marketing hype is over the top).

Quote
I am no colour guru but I know what looks 'good'. My point to Mike was that I thought DCF made normal images 'less good' by making the blue skies too mag. and the greens too yellow. Now it may be that this is more 'accurate', but I personally don't want 'accurate' I want 'good' - not the same thing at all.

Accurate is a buzz word that really yanks my chain. Accurate is by my definition colorimetrically correct (measured) color of the scene and by virtue the capture from a digital camera is scene referred. It will be accurate and quite ugly! Our jobs is to make output referred color which can't be accurate by my definition but can be pleasing or what YOU the image creator which to express of the image. Accurate color is often used to sell something. This was used early on in the color management days and maybe that's why the term upsets me because so few are willing to define it. It makes selling SOOOO much more difficult once really defined!

Quote
I was tempted to get out the GM chart but I thought what's the point? DCF does nothing for my files that I cannot do in PS and, on default, I don't like the effect.

I think most of us are in agreement on that!
Title: DCF Full Spectrum plug-in
Post by: paulbk on April 25, 2006, 08:21:57 pm
re: "... I don't like the effect."

I agree. I own DCF and think the hype and theory on tribecalabs' web site is, at best, misleading. If not 100% boloney. I thought Jon Cone was a color/print expert so it must be worth a try. Not necessarily so. DCF adds too much magenta to clear blue skies and green grass. See my examples page 2 of this thread, post #36.

Live and learn.

fyi.. Jon Cone's DCF page here (http://shopping.netsuite.com/s.nl/c.362672/sc.14/category.529/.f)
Title: DCF Full Spectrum plug-in
Post by: Nick Rains on April 25, 2006, 10:19:57 pm
Quote
Accurate is a buzz word that really yanks my chain. Accurate is by my definition colorimetrically correct (measured) color of the scene and by virtue the capture from a digital camera is scene referred. It will be accurate and quite ugly! Our jobs is to make output referred color which can't be accurate by my definition but can be pleasing or what YOU the image creator which to express of the image. Accurate color is often used to sell something. This was used early on in the color management days and maybe that's why the term upsets me because so few are willing to define it. It makes selling SOOOO much more difficult once really defined!
I think most of us are in agreement on that!
[a href=\"index.php?act=findpost&pid=63677\"][{POST_SNAPBACK}][/a]

I am actually far more interested in the way one sees colour and interprets vision than I am in technical colour stuff - although they are obviously related. it's a fascinating subject since it includes technology, psychology and physiology plus a bit of philosophy!

I, too, find the whole 'accuracy' concept flawed for real world applications such as selling nice photographs to hang on the wall and simply please the eye.
Title: DCF Full Spectrum plug-in
Post by: PeterLange on April 26, 2006, 01:31:27 am
By means of Calibrating Camera Raw:

“You’ll just get more accurate hue and saturation relationships in your images”.

Cited from: http://www.creativepro.com/printerfriendly/story/21351.html (http://www.creativepro.com/printerfriendly/story/21351.html)
by Bruce Fraser


Peter

--
Title: DCF Full Spectrum plug-in
Post by: Nick Rains on April 26, 2006, 02:21:49 am
Quote
By means of Calibrating Camera Raw:

“You’ll just get more accurate hue and saturation relationships in your images”.

Cited from: http://www.creativepro.com/printerfriendly/story/21351.html (http://www.creativepro.com/printerfriendly/story/21351.html)
by Bruce Fraser
Peter

--
[a href=\"index.php?act=findpost&pid=63699\"][{POST_SNAPBACK}][/a]

Sure, and your point is?

I have done the calibration as describe by Bruce and indeed find the results reasonably 'accurate' vis a vis the theoretical Lab values of the test chart. But the world is not a test chart and I find the results, applied to real world subjects, less than totally pleasing - subjectively of course.
Title: DCF Full Spectrum plug-in
Post by: PeterLange on April 26, 2006, 01:16:58 pm
Quote
I, too, find the whole 'accuracy' concept flawed for real world applications such as selling nice photographs to hang on the wall and simply please the eye. [a href=\"index.php?act=findpost&pid=63690\"][{POST_SNAPBACK}][/a]
.....

Quote
Sure, and your point is?

I have done the calibration as describe by Bruce and indeed find the results reasonably 'accurate' vis a vis the theoretical Lab values of the test chart. But the world is not a test chart and I find the results, applied to real world subjects, less than totally pleasing - subjectively of course. [a href=\"index.php?act=findpost&pid=63700\"][{POST_SNAPBACK}][/a]

My point is that there's no way to nice photographs without a system-immanent idea of color accuracy.  It’s hardwired deeply inside the input-pipeline starting with the sensor design.

Other people like Mr. Canon and Mr. ACR have already sorted this out for us behind the scenes (more or less). Even more, they already provide a first basic suggestion for a pleasing rendition on the top, output-referred, so that the user can focus on this last step to get everything ‘right’ within the limited dynamic range of the monitor and printer.

Calibration can definitively improve the communication between both gentleman mentioned above.  And yes, there are different concepts how this can be accomplished.

You may not believe in this line of thoughts, but on the other side if accurate color counts nothing, would you allow your children to paint new Bayer filters for your camera, or, let them randomize and tape down the Calibrate tab settings in ACR? I wouldn’t.

Sorry for leaving the original topic of thread.

Peter

--
Title: DCF Full Spectrum plug-in
Post by: bruce fraser on April 26, 2006, 02:50:02 pm
"Accurate" is not necessarily the final goal, but it tends to make for a much better starting point than randomly inaccurate....
Title: DCF Full Spectrum plug-in
Post by: digitaldog on April 26, 2006, 03:26:43 pm
In the case of the ACR calibration routine, you're taking output referred numbers in a defined color space and attempting to match those numbers to a target that would have those same values output preferred. So in this case, I think hitting those values (and I've never got dead nuts exact numbers but close), does assist in getting a good starting point as a new ACR default.
Title: DCF Full Spectrum plug-in
Post by: Nick Rains on April 26, 2006, 07:10:22 pm
Quote
.....
My point is that there's no way to nice photographs without a system-immanent idea of color accuracy.  It’s hardwired deeply inside the input-pipeline starting with the sensor design.

Other people like Mr. Canon and Mr. ACR have already sorted this out for us behind the scenes (more or less). Even more, they already provide a first basic suggestion for a pleasing rendition on the top, output-referred, so that the user can focus on this last step to get everything ‘right’ within the limited dynamic range of the monitor and printer.
--
[a href=\"index.php?act=findpost&pid=63753\"][{POST_SNAPBACK}][/a]

OK, I see what you're getting at, and I agree. Let me clarify my statements.

I feel that a pixel-peeping pursuit of pure accuracy is a flawed concept and will not lead to better pictures. Fujichrome was always a bit bright but it looked 'good' compared to Ektachrome which was arguably more accurate. Studio guys shot Ektachrome since they needed to match product colours, stock and landscape photographers shot Fuji because the colours looked better.

Having said that, as Bruce says, reasonably accurate colour is a necessary starting point but since it is such a hard thing to literally achieve, I rely on the camera to give me accurate enough colour and then I take it from there to the point which seems most satisfying. I have tried different methods to improve the basic colour response of the camera, ie calibration, and I have not found anything that meaningfully improves upon what the RAW converter is already able to do.

Which leads me back to the thread - plugins which base their approach on correcting real or imaginary inaccuracies in image capture are fighting an uphill battle. As Andrew pointed out earlier, once you define what you mean by 'accurate colour' you enter a very difficult region.

So, as long as Mr Canon has sold me a camera that gets me into the (small) ballpark I am happy. Somewhere in that ballpark is a distrubution of colours that I personally find appropriate and pleasing, whether or not those colours are strictly 'accurate' does not interest me.
Title: DCF Full Spectrum plug-in
Post by: PeterLange on April 27, 2006, 07:12:20 am
Quote
OK, I see what you're getting at, and I agree. Let me clarify my statements.

I feel that a pixel-peeping pursuit of pure accuracy is a flawed concept and will not lead to better pictures. Fujichrome was always a bit bright but it looked 'good' compared to Ektachrome which was arguably more accurate. Studio guys shot Ektachrome since they needed to match product colours, stock and landscape photographers shot Fuji because the colours looked better.

Having said that, as Bruce says, reasonably accurate colour is a necessary starting point but since it is such a hard thing to literally achieve, I rely on the camera to give me accurate enough colour and then I take it from there to the point which seems most satisfying. I have tried different methods to improve the basic colour response of the camera, ie calibration, and I have not found anything that meaningfully improves upon what the RAW converter is already able to do.

Nick,

Believe me, some problems on the way to get just nice photographs were the sole reason which brought me to all this technical stuff. Anyway, I hope that my posts did not damage the reputation of this discipline even more than some of the “science” which was offered related to the topic of this thread.

Referring to the term ‘reasonably accurate color’, it raises the fascinating question about the dosage of accuracy which a pleasing color should still contain. At the risk that it’s anytime possible to find examples for the opposite, I’d like to suggest a rule of thumb as follows:
Hue > Saturation >> Brightness.

Even a pleasing, output-referred rendition relies on a certain degree of Hue accuracy. Yes, there are tolerance ranges and a preferred reproduction of a specific color can involve hue shifts. But, my best guess is that strong memory colors such as skin tones are among the first who leave, if the camera + Raw converter fail to meet the hue of some related (red) patches on the ColorChecker chart.
Saturation seems to be of more volatile nature and pretty much subject of personal taste. Brightness can almost never be accurate due this whole subject of dynamic range compression (scene to output) and respective compensation measures (application of a sigmoidal tone curve).

So in full agreement with Bruce and Andrew that ‘reasonably accurate’ makes a better starting point - as well as in the sense that ‘options are good’ - let me refer to this former post here for some more details: http://www.adobeforums.com/cgi-bin/webx/.3bbf4d7a (http://www.adobeforums.com/cgi-bin/webx/.3bbf4d7a)


P.S.:  Looking at your website and in particular the breathtaking online gallery, what could I ever tell you about image processing & pleasing rendition…

Peter

--
Title: DCF Full Spectrum plug-in
Post by: tribecalabs on May 09, 2006, 05:01:40 pm
I’m with Nick, I am interested in the way one sees and interprets colors. It is a fascinating subject. (I referred to www.colorsystem.com earlier because it is a great history of color thinking, from Aristotle to CIE. If you enjoy color, you will enjoy this site.)

Claiming to improve the color accuracy of RGB is indeed an uphill battle, specifically because it raises issues of technology, psychology, physiology and philosophy. I hope that by providing a description of our method ab initio, as Peter requested, will give you insight to our philosophy as well as clear up any lingering misunderstandings. Because while it is true, there is nothing wrong with your digital camera, but it is also true that there are limitations to the RGB model used to render digital images.

To begin, as Nick points out, it is important to define what one means by accuracy. My definition of color accuracy is simple; the colors of an image should be visually consistent with the colors of the input. This is the basic philosophical difference between my definition of color accuracy and those that define accurate as a colorimetrically correct. Accurate, to me is a perceptual match based on visual analysis of image and input. Or simply: the picture should look like the thing.

As I stated earlier, my experience with digital photography is primarily in the field of fine art reproduction, where color matching is critical. I have had the benefit of shooting thousands of high resolution digital images in ideal shooting conditions, with the highest quality digital cameras and the luxury of comparing the image on perfectly calibrated equipment to the original art under the very light with which it was recorded. Through the course of my work with the Herbert F. Johnson Museum of Art at Cornell University 1998-2000, I began to notice inconsistencies between image and input. Curiously, the hundreds of thousands of dollars worth of equipment could not seem to reproduce certain colors accurately (by my earlier stated definition, some colors on screen did not visually match the colors of the original object). This was despite the efforts to properly expose and balance the image and viewing the image on a calibrated system. The difference was most notable in the deeper colors, such as violet and indigo.

As part of my job description, I was responsible for making color accurate reproductions and since this was new technology, I had to do some research as to why my recordings were not color accurate. The first step was to confirm my observations. The first confirmation of the color disparity that I found was by Robin Myers, in his article, “Color Accurate Digital Photography of Artworks” in which he observed the difficulty of reproducing cobalt blue colors (similar to my problem with deep violets and indigos). Since he is the creator of Colorsync, I knew I could take his word as a reliable source.

Later working as a technical representative for a European medium format camera-back manufacturer, I continued to observe the chromatic disparity between certain hues and their digital reproduction. This was problematic for me, because it was my job to introduce extremely expensive digital camera-backs to professional photographers shooting catalog work, and as we all know, one of the main reasons for returning merchandise bought online or in a catalog is that the color doesn’t match. As Mr. Nemo says, it may be true that in print production, the prepress house has the final say on color matching, but the photographer is the first step in the chain of production and often the only one with the actual article in front of him for comparison. Being that I was trying to sell a very expensive piece of photographic equipment, the photographers were very critical of the results on screen (yes, calibrated). So my research into the color problem continued out of necessity.

I found more confirmation of my original observation of the color differences between input and image by industry leaders (whom I have cited in earlier posts) including Robin Myers, Michael Stokes, and Charles Poynton. I observed a phenomenon, the disparity between image color and original; researched it to confirm my observations; and came up with a hypothesis as to why the phenomenon occurs.

As Mr. Rodney mentioned earlier, digital cameras do not produce color; they simply count (quantify) light input values. These input values, recorded spectral power distributions, are mapped to RGB tristimulus values using CIE colorimetry.

The hypothesis: Although CIE colorimetry is necessary for communicating color data between devices that produce colors, CIE colorimetry alone is not sufficient for characterizing light input data (camera file) such that a resulting RGB image is consistent with human color perception of the scene recorded. In other words, the picture doesn’t match the thing: colorimetry alone doesn’t fix the problem.

As to why this may be the case is the subject of color theory. All CIE colorimetry is based on the famous 1931 color matching experiment which was designed to test the laws of trichormaticity. It was believed, and rightfully so, that a minimum of three colors can be combined to form all colors. In the experiment, the intensity of three energy constant (single wavelength) lights was adjusted so that their combination visually matched a white reference light. Colors in this model are not based on the way we see colors, they are fractions of the whole, white light. CIE colors are mathematical abstractions plotted on a two dimensional graph. This graph serves as a very useful tool for comparing two measured colors between devices that produce colors, but as Mr. Rodney points out, digital cameras do not produce colors.

RGB is the model used to produce digital colors. RGB is an additive model of illumination using three colored stimuli. You can clearly see how CIE colorimetry can be useful for providing a white point (color temperature) since that is almost the color matching experiment to a tee and for determining the gamut of the RGB model by specifying the primaries; however, the RGB model is a model for producing illumination, not for characterizing human color perception and RGB illumination is very different than most other light we see.

Just as colors are a mathematical abstraction in CIE, colors of RGB illumination are equally spaced fractions of white light. I believe this is the difference between RGB and human color perception. Colors of RGB illumination are additively combined, whereas we see things by the subtractive process of absorption and reflection of light, or in the case of the sky, the process of refraction. In these two examples, colors are not equally spaced fractions of the whole as in RGB. There are color differences between two models: RGB illumination and the human response to what we can call for lack of better terms, real illumination.

RGB illumination breaks down in chromatically consistent intervals, meaning as an RGB color is reduced in value, the percent R:G:B remains constant. Real illumination breaks down in a very nonlinear manner. This may be why objects get deeper in appearance as they get darker, (unlike the colors of RGB). Just as the CIE experiment used energy constant light sources, RGB illumination is modeled to be independent of intensity; otherwise your computer screen would get hotter as it gets brighter. Most color we see, and therefore record with our digital cameras, are caused by light/heat energy, and as the temperature of the source of the illumination changes so to does the chroma of the hue illuminated.

We theorized that by varying the chroma of the hues of the RGB color model relative to their intensity, we could produce RGB images that were visibly more consistent perception of real illumination.

To test this theory, we set up a full blown psychophysical evaluation of color difference between recording, the image, and recorded, the object. We took great care to eliminate as many environmental variables as possible, such as the ambient light of the surround, and made sure to use equipment both in its default state and in a state calibrated to ISO standards. We identified commonalties in the color differences using multiple devices. We then verified that our observations were consistent in random tests, not subject to strict imaging guidelines.

This work led to a comprehensive characterization of color difference between RGB and human color perception. This characterization of color difference serves as the basis for our color difference model: the formula, which we have called DCF Full Spectrum. (A color difference model is a color appearance model that attempts to minimize the perceived color differences between two color systems.)

As has been pointed out in the course of this discussion, we enacted changes the RGB model using an image editor, such as Adobe Photoshop (so that we could offer our results as a Photoshop plug-in). The result of our color difference evaluation is a series of value modifications that modulates the chroma of RGB hues relative to their intensity. The tools we use to enact color changes, like any automate plug-in, are less important than the formula itself. The goal of DCF Full Spectrum is to produce a more accurate recording. Making a picture pretty so that you can hang it on the wall is still subject to your skill set and your aesthetic decisions (just like choosing Kodak film over Fuji film does not automatically make you a better or worse photographer).

My partners and I are very proud of our work and we are happy to offer our DCF Full Spectrum color model to digital photographers. If you have any questions please feel free to contact me at contact@tribecalabs.com.

Thank you,

Michael Bevans
www.tribecalabs.com
Title: DCF Full Spectrum plug-in
Post by: jani on May 11, 2006, 06:29:30 am
Mr. Bevans,

Wouldn't it have been far easier for you to implement this as your own RAW converter instead of relying on a RAW converter's interpretation of the image source before you manipulate your colours?