Fooling your camera?

[Ant Carlos]

I'll start this new topic, although it is related to most of the old ones bellow, there's a new thing I would like to discuss separately. And I would certainly love to hear Mike's opinion.

As many of you, I work with long distance commissions and use photos to discuss the WIP with my clients. One thing that keeps bothering me is that the photo of my work never looks like the real one. Not even close. I try to adjust colors etc., of course, but it always seems to be worse than the painting in front of me. I assume people who work Alla Prima may have less problem, if at all. But my paintings are built with layers, with lots of transparency. To naked eye it makes very good 3D effects, which are all lost in the photos. I am pretty sure the cause is that the camera sees what our eyes cannot see. Like astrophotos that show colors in the sky, the camera shows some undesirable colors in my paintings. Sometimes subtle ones, but it really bothers me. So yesterday I did a test and put my hand in front of the painting I am currently working and took a shot. Somehow the presence of my real flesh tones made the flesh in the portrait look more natural. I started to believe that the CCD of the camera calculates an average when shooting. So when you shoot only the painting it calculates every spot of colors that it sees (remembering it sees much more than our eyes) and produces unreal results.

OK, I know it sounds a bit confusing so let me show you what I am trying to bring out. The first image is a photo of the painting in progress (the best I could manage to get) with it filling the whole field at the moment of the shot. The second one was taken with my hand in front of the painting. See if you find how the 2nd one looks more natural.

http://antcarlos.com.br/upload/regina-wip5.jpg

http://antcarlos.com.br/upload/regina-compare.jpg

So now I wonder if this is a nice method to fool the camera when taking pics to discuss with clients. Of course I'll have to go through other tests, but I am sharing with you. Maybe someone out there has the same problem.

Ant

[Michele Rushworth]

They look the same to me.... both are magnificent. I know what you mean that the photos never look as good as the painting, though, with lots of color distortion. I think every artist has that problem.

[Richard Budig]

Ant . . . I've been painting posthumous portraits for several years, so I have the same set of problems you mention. Over the years, I've learned that cameras, film, digital equipment does what I call "value clumping." I heard that term from Chris Sapper on this forum quite some time ago. Thus, it puts the responsibility on you to interpret things . . . is it really that dark in there . . . are those lights really that "blown out?" Generally, you'll improve your work if you flip those things a little bit . . . lighten some of those darks, and darken some of thoss lights. For example the darkening value separating the jaw from the neck is often much too dark in a photo, and often, the highlight areas are valueless white blotches. All that being said, in the end it is pretty much up to you to make sense out of what are sometimes rather poor photos resources. In my case, I paint young men and women killed in the war, so many of my photo resources are those typical little grab and shoot photos taken while he/she was home on leave. In many cases, I find that all I can do is use the photo as a guide. Sorry I can't be more specific.

[Mike McCarty]

Ant,


They look very similar to me as well. Your hand looks a bit lighter, but most hands, and undersides of our arms will look more pale than the top which gets more sun. Also, your hand is catching the light the way any 3D form would, and unlike the flat surface of the painting.

I could see maybe if you were offering a crop of a very small portion of the painting that the camera might try and interpret multiple layers and even pick out isolated coloration. The camera, if it's a good one, is only doing exactly what your want it to do - render as faithfully as it can what is in front of it. However, when you back up to this distance there shouldn't be a problem if your procedure is sound.

There are some good threads on photographing art work in the building next door. It can be a daunting task, indeed. But truly, you seem to be doing a pretty good job of it.

It seems to me that if your client is getting this kind of reproduction of a wip they could not hope to get much more.

As the queen once said: You may be protesting too much, methinks.

[Mike McCarty]

Ant,

If I could offer an addendum ...

One thing that can bugger up the color of your photograph is the "white balance" setting on your camera. Check to see that the light hitting your painting is the same light your camera is expecting, such as: incandescent, flourescent, natural shade etc.

However, I'm not sure that I am getting to the core of your question.

Your camera's sensor will "meter" the light usually in one of three ways depending on how you've set it. There is "matrix," which is a total image calculus, a "center weighted," and a "spot" metering, which calculates it's exposure based on a small portion of the image that you have selected in your viewfinder.

If you introduce your arm into a "matrix" metering I think it would alter the calculus somewhat depending on how much of the frame you replace. This could alter the total image calculation / exposure such that the overall could look lighter or darker.

I would say that these meter calculations are basing their findings (exposure) more on light than color. This gets a little fuzzy. Whereas light reflects off of one color differently than another, but I think it's all about the amount of light and not color.

I hope I haven't muddied the water further.

[Terri Ficenec]

I've also found that colors and contrast are skewed when taking artwork photos. When photographing subtle colors, the camera seems to need to spread things over the full range of colors and values. So I'll include a color range and something black and white along side the painting to prevent the camera from exaggerating what should be subtle differences.

The two images of the WIP watercolor below were taken in the same light, same camera settings a few moments apart from each other... only difference is what else besides the artwork is included in the photo. To me, the one with the multi-colors and white/black is a much closer representation of what the piece actually looks like than the version with the dark only background which results in overblown lights and too much contrast where there should be soft color transitions . .

[Ant Carlos]

Mike, thank you.
You are always ready to discuss those kind of issues and I appreciate that.
I understand your point and no way you made the water muddier. Actually I think you explanation is very clear and may be close to the answer. What Terri shows us is an ilustration of what you said. When she added the whole spectrum to the field her camera created a much more balanced picture of her work. It's very like what I did when I put my arm in front of the camera and provided more 3D shapes and colors for it to deal with. I found that resulted in a more natural representation of my painting.
Anyway, what I really wonder is if the camera capability of seeing colors out of range of the human eye doesn't play a role in this act. I refer again to astrophotos (I am an amateur astronomer myself and had the chance to testify several times that what our eyes see are much less than what the camera sees and register). Just google for some shots of the milky way. You'll see many colors that your naked eye can't see. So my question still remains: is the camera showing some frequencies of liight that we painters cannot spot, so we always find our photographed works uglier than the original?

[Mike McCarty]

So my question still remains: is the camera showing some frequencies of liight that we painters cannot spot, so we always find our photographed works uglier than the original?

To tell you the truth - I don't really know.

There are a couple of truths that we won't be able to get around. One: we're stuck with our eyes. Two: we're stuck with the current iteration of camera technology, which by the way is truly remarkable.

When I take the final, final photo of my artwork I want it to be as dense with pixels as I can make it. This means that I want to fill the frame only with my painting to the extent that I can. I don't want to waste any pixels on the wall behind, or any other artifacts. This, so I don't have to crop some portion of the photo image away, thus leaving pixels on the table, so to speak. If, because the difference between the shape of my painting and the shape of my photo image is such that I have a portion that must be cropped out, then I want to fill that space with a matte black cloth. This method has given me my best color reproduction. Beyond that, if my camera is good and if my procedure is good then any variations from what is real and what is depicted should be small, difficult to measure, and probably insignificant in the grand scheme of things.

[Ant Carlos]

Mike,
I think you can better see the undesirable spots (frequencies of light?) In this close up: http://www.antcarlos.com.br/upload/r...s5a-detail.jpg
The photo shows some colors and details that are not so prominent in real life. I think the oranges and violets seem to come out way to much. The tiny hairs are too hard in the photo, opposite to the real painting in which you can barely see them (please go back to my first post and look the photo taken with my hand in front of the painting. See how her face looks smoother with better transitions than in the other photos). Yeah, I know I may be asking too much from the digital world, wich I agree with you is remarkable nowadays. I just started this because I'd like a representation of my painting to be more real and somehow I believed the fact of adding a real skin tone in front of the portrait made it look closer to the way I see it.

[Mike McCarty]

Ant,

I did go back and look again and I do see a difference in the two images. The one without your hand seems a tad lighter. It's hard to be definitive given the variances in color representation between computer monitors and even the possible variance in lighting between these shots. Unless you are using artificial lighting the conditions of light can vary from second to second.

I don't think I'm going to be able to satisfy your question except in the ways that I have previously tried. Maybe if you could tell me what kind of camera and light set-up you are using I might be able to investigate further.

Even if you are right you can't arbitrarily stick your arm into every photo you take. That might be a bit more distracting than the conditions you are complaining about. Ultimately we are left with human eyes that look upon things that are mechanically reproduced.

Perfection is elusive and rare, my southern himispherical brother.