Date   

Re: White Balance Dilemma

David Lawrence
 

Thank you Henry!


I'm beginning to find my way back to the light. :)  We have been fighting this screen calibration for so long that I forgot some basic tenants of white balance and PPW. I found this explanation useful to *balance* my thinking again.  I think Dan correctly pointed out that this scene is pretty "wacky" and everyone has highlighted the flaw in my approach.  I completely lost track of the intended subject of the photograph as I focused solely on the color of the background lights. 


"Color Temperature Explained

Before you go into the process of correcting color imbalance you will need to understand color temperature. A basic description of color temperature is based on the color characteristics of visible light from warm (yellows) to cool (blues) and the ability to measure this in degrees Kelvin (°K). Degrees Kelvin is a numerical value assigned to the color emitted by a light source. Visualize a lamp filament that is heated using an electric current. It starts off as black and starts getting hot. At a particular point it will become hot enough to start glowing, typically a dark red. As it gets hotter, it will change from dark red to orange to yellow to practically white. It is important to understand that technically, red light has a lower color temperature but is described as warm, while blue light is a higher color temperature but is described as cool. So remember that the terms warm and cool describe color, not temperature. This is a fairly extensive topic but for a quick explanation this should help."


Balancing Color for Flash and Ambient Light using Gels

 

David Lawrence


Re: White Balance Dilemma

Henry Davis
 

Please don't deduce anything because I'm just writing by guesswork. I was suggesting that synthetic sensors don't react exactly the same as human sensors. I believe that emitted and reflected components are handled differently by humans and cameras - it's just a hunch. I'm suggesting that reflected light from a given light source differs by magnitudes as that light source is changed - as the human experiences it. It's not the same magnitude for a mechanical sensor.

When you view a scene do you really consider the light source? Photographers probably add an understanding into their viewing experience but even with them the subject is the most important component. The reflected light is the main concern - is it not?

As viewers our eyes and brains consider a scene and we make up a composite mental scene - the "final". The light source is secondary to the experience - even though it has a primary effect on the initial perception. Cameras shouldn't be equivocated with human experience even though they seem to mimic it in an astounding way.

Simultaneous contrast is a fact of life - synthetic sensors have no perceptual cognition, thus the human experience of simultaneous contrast is perturbed by comparison. There's no one-for-one in mechanical and human visual "experience". The human experience is a different thing than a simple mechanical response.

The difference in human and mechanical "perception" - especially between emitted and reflected light - are not the same and I believe that an equation for compensating the two isn't in the cards. Maybe close but not the same.

My short answer is that you do not primarily consider the light as it is emitted. Your first consideration is the scene. But I'm just guessing.

Henry Davis

On Apr 4, 2016, at 6:57 PM, david@pixelpurfect.com [COLORTHEORY] wrote:


Interesting Henry, should I deduce from that; my eye sees the light as emitted and my camera sees it as reflected?



Thanks,



David Lawrence


Re: White Balance Dilemma

Dan Margulis
 

David Lawrence writes,

Thank you Jim, those images are very useful, what I am stumped by is why when the white balance is correct the upper spot lights appear to be blue, yet to the naked eye they are clearly magenta/purple.

If you shot directly into the sun you would also be unable to get a proper balance, for the same reason: cameras have no way of recording colors that are simultaneously extremely vivid and extremely light—they’re way out of the gamut of any RGB, even ProPhoto. The result is an unnatural look when the light source hits a darker area (where the camera *can* accurately report color). A hue change in the light source can’t be unexpected.

It would be bad enough if the light source was all of one hue, but here you have multiple competing casts. Correct one, and you make the others worse.

We humans adapt to multiple casts much better than a camera can. When the lighting is this wacky, of course we don’t see the scene as neutral. If the light sources were less intense, then we’d just see less of the cast(s) than the camera does. But here the situation is weirder: we see less color than the camera does in the midrange but *more* at the endpoints. Translation: there is no easy solution. To get a believable result requires improvisation.

Two suggestions based on my experiences with this type of image (it comes up a lot, because it’s very frustrating).

*The H-K action works very well with this kind of scene. It damages all colors, but the more colorful the original object the less it gets hurt. So, after running it, you will probably have to run something like Color Boost to regain the original brilliance in the lighting itself—but everything that is not one of these brilliant light sources will be toned down.

*We know that we would never perceive the performers and stage objects as they would appear under sunlight. Nevertheless, it pays to, very quickly, cobble together a version of the image that does exactly that. Of course, it will look ridiculous if we have “normal” fleshtones when the lighting is this bizarre. But it’s quite useful for blending. We know we have to go *somewhat* in the direction of more normal lighting, but it’s hard to know how far. Putting this bogus version on a layer and playing with opacity will answer that question.

Dan Margulis


Re: White Balance Dilemma

David Lawrence
 

Interesting Henry, should I deduce from that; my eye sees the light as emitted and my camera sees it as reflected?


Thanks,


David Lawrence


Re: White Balance Dilemma

Henry Davis
 

Getting the lights to your liking using global corrections would ruin the scene. If you dislike the lights handle them separately.

Your confusion is likely caused by the real difference between reflected light and emitted light. Maybe?

Henry Davis

On Apr 4, 2016, at 5:26 PM, david@pixelpurfect.com [COLORTHEORY] wrote:


Thank you Jim, those images are very useful, what I am stumped by is why when the white balance is correct the upper spot lights appear to be blue, yet to the naked eye they are clearly magenta/purple.


Re: White Balance Dilemma

David Lawrence
 

Thank you Jim, those images are very useful, what I am stumped by is why when the white balance is correct the upper spot lights appear to be blue, yet to the naked eye they are clearly magenta/purple.


David Lawrence

www.dllawrence.com


Re: White Balance Dilemma

Jim Lawson
 

https://www.dropbox.com/s/xycp5nbtt7ixv0c/DSC00103.jpg?dl=0

I Basically agree most of the observations. I brought your raw files into a ACR and just balanced with the eyedropper tool on the black T-shirt of the lead singer that straightened a lot of things I think. As long as he looks reasonable the rest of the lighting makes sense.   If you want to download the JPEGs i created you have the links.  

Jim Lawson
                                            
website                        http://www.jimlawsonphotography.com                                    
blog                            http://jimlawsonphotography.wordpress.com/       




On Apr 4, 2016, at 4:10 PM, david@... [COLORTHEORY] <COLORTHEORY@...> wrote:


Actually it does help Kirk.  For some odd reason I had in my mind that the "color temp" of the bulb/source would remain the same when a gel is placed over it.  Not sure how that got in there, but I believe I'll show it the door.  :)


I will take some time and adjust the image WB in Lightroom and see how it shakes out. 


Thank you!


David Lawrence

www.dllawrence.com




Re: White Balance Dilemma

David Lawrence
 

Actually it does help Kirk.  For some odd reason I had in my mind that the "color temp" of the bulb/source would remain the same when a gel is placed over it.  Not sure how that got in there, but I believe I'll show it the door.  :)


I will take some time and adjust the image WB in Lightroom and see how it shakes out.


Thank you!


David Lawrence

www.dllawrence.com


Re: White Balance Dilemma

kirkthibault
 

Hi David,

Without really knowing what colors things in the setting are supposed to be, and noting that the lead acoustic guitarist and the accompanying female singer appear to be the most “important” subjects in terms of the lighting, my inclination would be to make their skin tones look good.  

While there is a bunch of colored downlighting (is that the correct term?) coming from above and behind the band and the subjects, there is tungsten-like lighting coming from in front of the stage (like spot lights from the back of the auditorium) that are illuminating the two main subjects.  You can see the shadows falling on the lead guitarist’s legs from the music stand and mic stand coming from this lighting.  I believe that the first shot, the one with the guitarist with his hand up and the female singer sitting down, has an “As Shot” WB in Capture One of about 3400 K and a slight (-3.3) tint toward green.  This makes the skin tones of the two main subjects look more or less “accurate” and pleasing enough.  If you copy and paste this WB into the second picture, you get similar skin tones and you can appreciate that the background and stage lighting has changed.  

Thus, the main lighting is coolish tungstenish, 3400K lighting from spots illuminating the two main subjects.  I’d go with that, and bring the exposure down slightly (about half a stop).  ACR says the WB is 3250K in the first image and 7700 in the second image, so similar enough.  

See if applying the As Shot WB from the first image to the second image straightens things out.  The second image has an As Shot WB of about 7600K in Capture One and the whole scene appears way too warm.  The chandelier lighting above the stage in the second image also causes spill of that warmer light onto the stage and the seating in the auditorium, further making the very high, overly warm WB color temperature really permeate the image.  The skin tones for the two main subjects are too intensely warm as well.

These images contain lighting that is artistic and theatrical and that is the color of the light - trying to counter the intense magenta, for example, will turn the image greenish - yow!  WB for the two main subjects in the tungsten-like light and let the rest fall where they may.  They are getting bathed in that theatrical colorful light.

Hope this helps.

Kirk Thibault




On Apr 4, 2016, at 2:33 PM, david@... [COLORTHEORY] <COLORTHEORY@...> wrote:


Absolutely Kirk, here is a drop box link to the Sony RAW files.


Color Theory Group


Thanks,


David Lawrence

www.DLLawrence.com




Re: White Balance Dilemma

David Lawrence
 

Absolutely Kirk, here is a drop box link to the Sony RAW files.


Color Theory Group


Thanks,


David Lawrence

www.DLLawrence.com


Re: White Balance Dilemma

Gerald Bakker
 

You say, the stage lights are 3200, that's a warm yellow color. Illuminate a scene by such light and everything gets a yellow cast.
A custom WB set to 3200 will make your camera add a lot of blue to the whole image, to neutralize this yellow cast.

The problem with the magenta spots is that they should not be corrected like this. They are magenta and not really colored by the yellow stage light. Yet, the WB 3200 setting will make them blue like everything else. That's what you see in the first photograph.

The difficulty of your example images is that they are so different. As the originals are RAW, why not pick one, set the color temperature to 3200 and 7700 resp. and export both results?

I believe a scene like this is very hard to get perfect with just one WB setting. There are multiple, differently colored lights, so everything is colored by some mix of yellow, blue, magenta and whatever color.
The best advice is to just use a fixed WB setting and correct in LR or CR by the WB sliders until it looks good. Some local correction may be necessary though.

Gerald Bakker
Gerald Bakker - Home

 


Re: plugins

Jim Donovan
 

I will second Doug Schafer's suggestion of using Lumenzia. Outstanding plug in, well worth the price. There are a ton of instruction video's explaining all the tools.One of those tools the more you play with the more useful it becomes. Use it daily and customer service from Greg Benz was stellar. Jim Donovan

On Mon, Apr 4, 2016 at 10:00 AM, ds-mail@... [COLORTHEORY] <COLORTHEORY@...> wrote:
 

Also consider using Lumenzia


Doug Schafer



Re: plugins

onelistdrs
 

Also consider using Lumenzia


Doug Schafer


Re: White Balance Dilemma

Kirk Thibault
 

The two images you posted have pretty different lighting conditions in them - for example, the second image that you WB’ed around 7700K has a lot of mixed, what appears to be tungsten-like lighting in the central chandelier fixture, in addition to the filtered stage lighting.  

Any chance you can post links to download the raw files?

kirk


On Apr 4, 2016, at 8:09 AM, david@... [COLORTHEORY] <COLORTHEORY@...> wrote:


Hi Everyone,


I have a white balance dilemma I would like to run by minds brighter than mine.  Our church recently installed new video cameras and three 50 foot LED panels.  We have always had a problem reproducing magenta onto the screens.  We have white balanced everything many times and I'm not looking for a fix for that.


I happened to be taking photographs during rehearsal as I often do and I noticed the lights on stage where showing up on my display as blue when they appeared magenta, (to my eye).  I was shooting with Auto White Balance, with the latest Sony a6300 sensor. I decided to switch to manual white balance and was surprised by the results.  I am told our stage lights are around 3200.  I uploaded a couple of my pics to the link below. The first was shot at 3250 and the top two lights appear blue, the second was shot at 7700 and they appear magenta, as seen by myself and others in the room. Shouldn't the color be correct when the color temperature matches?  The jpgs are from RAW files exported straight out of Lightroom.  Any ideas?  I just don't get it.


Applied Color Theory in Photoshop

 

Peace,


David Lawrence

www.DLLawrence.com




Re: White Balance Dilemma

Henry Davis
 

If the goal is to color correct the scene (not to reproduce it as it appeared to an observer) the white point is only one end of the tone scale that will need to be addressed. Some attention will need to given to the entire tone scale.

A "custom" white balance isn't arrived at the same way as a "manual" white balance setting. They aren't the same thing.

Light bulbs or lamps are poor target objects for correction. When making a custom white balance avoid using the light source as the target. A test shot using three grey cards of three different tonal values might be helpful as might a grey ramp. The in-camera custom white balance will be somewhat different for each grey or "white" value. You can choose the custom white balance that gives the best overall results or the one that gets it better for the portion of the tone scale that is most important.

Again, you might need to clarify the goal of the "color correction" - I may be misunderstanding your post.

Henry Davis

On Apr 4, 2016, at 8:09 AM, david@pixelpurfect.com [COLORTHEORY] wrote:


Hi Everyone,



I have a white balance dilemma I would like to run by minds brighter than mine. Our church recently installed new video cameras and three 50 foot LED panels. We have always had a problem reproducing magenta onto the screens. We have white balanced everything many times and I'm not looking for a fix for that.



I happened to be taking photographs during rehearsal as I often do and I noticed the lights on stage where showing up on my display as blue when they appeared magenta, (to my eye). I was shooting with Auto White Balance, with the latest Sony a6300 sensor. I decided to switch to manual white balance and was surprised by the results. I am told our stage lights are around 3200. I uploaded a couple of my pics to the link below. The first was shot at 3250 and the top two lights appear blue, the second was shot at 7700 and they appear magenta, as seen by myself and others in the room. Shouldn't the color be correct when the color temperature matches? The jpgs are from RAW files exported straight out of Lightroom. Any ideas? I just don't get it.


White Balance Dilemma

David Lawrence
 

Hi Everyone,


I have a white balance dilemma I would like to run by minds brighter than mine.  Our church recently installed new video cameras and three 50 foot LED panels.  We have always had a problem reproducing magenta onto the screens.  We have white balanced everything many times and I'm not looking for a fix for that.


I happened to be taking photographs during rehearsal as I often do and I noticed the lights on stage where showing up on my display as blue when they appeared magenta, (to my eye).  I was shooting with Auto White Balance, with the latest Sony a6300 sensor. I decided to switch to manual white balance and was surprised by the results.  I am told our stage lights are around 3200.  I uploaded a couple of my pics to the link below. The first was shot at 3250 and the top two lights appear blue, the second was shot at 7700 and they appear magenta, as seen by myself and others in the room. Shouldn't the color be correct when the color temperature matches?  The jpgs are from RAW files exported straight out of Lightroom.  Any ideas?  I just don't get it.


Applied Color Theory in Photoshop

 

Peace,


David Lawrence

www.DLLawrence.com


Re: plugins

Jacek Poplawski
 


Re: plugins

Ronny Light
 

Jacek,

 

 

I’m a long-time user of what is now called the Nik Collection.  I love the collection and will continue to use it.

 

In the last few months, I have spent a lot of time learning and using luminosity masks.  I think Tony Kuyper’s TKActions and Sean Bagshaw’s videos about TKActions are wonderful.

 

Like you, I don’t limit TKActions to landscapes.  I use them on many types of images.  I use the panel for Photoshop CC 2015.

 

If you haven’t already, you should check out the Channels Power Tool at KnowHowTransfer.  Without leaving RGB, you can generate masks for all of Dan Margulis’ 10 channels – RGB, LAB, and CMYK.

 

From: COLORTHEORY@... [mailto:COLORTHEORY@...]
Sent: Saturday, April 2, 2016 4:56 PM
To: Derick Miller derickdmiller@... [COLORTHEORY] ; COLORTHEORY@...
Subject: Re: [COLORTHEORY] plugins

 

 

Another post I agree :) 

 

I use TKActions for every photo (and I do portraits, not landscapes). I know one can generate luminosity mask in PS but the power of actions is that you know what you can expect and you can concentrate on creativity instead repeat same steps. 

 

BTW I tried to explain to TK how same technique can be used in LAB to generate "pseudo luminosity masks" based on A and B and he said he will introduce this idea in future versions. Masks generated from LAB are very useful so automating it similar way will be awesome.


Re: plugins

Jacek Poplawski
 

Another post I agree :) 

I use TKActions for every photo (and I do portraits, not landscapes). I know one can generate luminosity mask in PS but the power of actions is that you know what you can expect and you can concentrate on creativity instead repeat same steps. 

BTW I tried to explain to TK how same technique can be used in LAB to generate "pseudo luminosity masks" based on A and B and he said he will introduce this idea in future versions. Masks generated from LAB are very useful so automating it similar way will be awesome.


Re: plugins

Derick Miller
 


U-point technology is very good at generating quick, localized corrections which look organic. They utilize a combination of tonal and color information within the user defined and quickly manipulable area. Even a very skilled hand would have a hard time doing it nearly as quickly in PS. 

As my PS skills have improved, I rarely use NIK, because it takes time to launch and get back out of NIK and I am more able to get what I want done. But there are types of selection it is still much better and quicker at achieving. 

I find TKActions quite helpful for speeding up my process and combining it with the basic selection tools, apply image and calculations in PS gets me much of what I want to accomplish. 

If PS could license U-point, or if someone could build a third party solution which could generate selections using that technology inside PS, I would value it. It would speed my process in some cases. 





On Apr 1, 2016, at 16:56, Jim Goshorn jgoshorn@... [COLORTHEORY] <COLORTHEORY@...> wrote:

 


On Apr 1, 2016, at 4:23 PM, Jacek Poplawski jacek.poplawski@... [COLORTHEORY] <COLORTHEORY@...> wrote:

I also disagree that Nik control points are worthless and you can do same in PS. Sure you can do everything in PS, including paint whole picture from scratch, but I don't know way to emulate control points in pure PS without Nik. This is not just luminance mask. This is special algorithm which probably uses some kind of segments to prepare good mask.

There is a way that allows you to toggle the control point mask display on and off and if you display the mask, you will discover it is not that accurate. Many times, it is not that important but if you really need a _good_ mask, you are better off doing it in Photoshop yourself. Yes, you can use minus points, but sometimes that can be more problematic than it is worth.

For the most part, I gave up on U-Point a while back.

Jim

3861 - 3880 of 33218