Fun with a model! How bad can it look?

Hi, Will,

Will Thompson said:

Here we go!

EF 50MM f1.2L Prime 1/60 f4.0 ISO 400 1DsMKIII MR14EX Ringlight​

Click to expand...

Precious! The model is exquisite and I like the pose very much.

I think I'd like to see a little more contrast - the image conveys a slightly "faded" look.

Was the ring light the only lighting? (Doesn't seem like it from the catchlights.) I have used a ring light for informal portraits, and was often disappointed in the overly "flat" result vis-à-vis even a small softbox on a shoe-mounted flash unit.

Best regards,

Doug

Will Thompson said:

Here we go!

EF 50MM f1.2L Prime 1/60 f4.0 ISO 400 1DsMKIII MR14EX Ringlight​

Click to expand...

Hi Will,

I like to see you are out shooting again! How on earth did you match those colors!

I see good and bad in that, but it comes out ahead and is a simple design. This girl has a relaxed non pouting pose that works. That bench might be awkward for her with the divider in the middle to stop the transients from bedding down for the night. Still, the extra steel is no match for her agile long legs. With her, you'd never realize that! The ring light s exceptionally good for opening up shadows but then the risk is less dimension. Here it works just fine.

The great advantage of your setup is extreme portability with nothing to have to lug around or setup.

Asher

Hi Will, I hope you don't mind some comments from the peanut gallery.

First, congratulations on having a gorgeous woman in front of your lens. I find shooting people challenging. So you already have my admiration.

Where I have some constructive comments are as follows:

1. Skin - looks overly processed (too smooth) and color seems off
2. Eye whites are quite yellow
3. Excessive contrast between hair and face
4. Amputee syndrome

Skin is always a touchy subject. Some people like Barbie like skin--smooth and unrealistic. Others prefer a more modest touch. I fall into the latter camp.

While eye whites are not truly neutral, they are not far off. The model's eye white are quite yellow.

The contrast between face and hair seems quite strong, though that is likely the style you were aiming for.

Amputee syndrome refers to when you crop above the joints. Usually, when you crop a person, you want to crop above the knees and above the elbows. Otherwise, you have an amputee look.

When I shoot people, I try to run my eye quickly around the box, looking for cut body parts.

An internet friend, who is a well known pretty girl shooter in L.A., recently wrote an e-book called "Guerrilla Glamour." His book is aimed at those who enjoy shooting pretty girls as a serious hobby or as a profession. He teaches you that you don't need the latest and greatest equipment, and where you should focus your precious capital. He also covers how to work with models, pose models, some general guidelines and more. His e-book spans over 100 pages, as I recall, and most of it is meaningful text. In other words, it isn't simply eye candy. I highly recommend investing ten dollars to read his book.

As as disclosure, although I have never physically met the author, he and I correspond frequently by email on a number of varied topics. I enjoy his intellect and insight. Although we correspond and I hope his book does well, I get no benefits from mentioning or promoting his book, other than good karma.

She's a pretty model and you did well to have her pose for your shot.

To all:

This shot turned out just as intended with no editing and custom white balance set on location with a gray card.

Doug:

The only other lighting was the morning sun reflected off of the next building.

Will Thompson said:

To all:

This shot turned out just as intended with no editing and custom white balance set on location with a gray card.

Click to expand...

I often use LAB mode to quickly scan a photograph. Looking here, her eye whites show a slight green bias and heavy yellow. Eye whites should be neutral or warm colors--that is, slight magenta (positive a) and slight yellow (positive b). As we age, we get further away from neutral. +14 does seem a bit high.

Still the color is accurate!

Will Thompson said:

Still the color is accurate!

Click to expand...

Will,

There's no criticism of your picture. So if it's accurate and one finds, as here, perhaps more yellow than expected, then what might such color due to? Is she asian, (not sure that sclera of Asians are more yellow) or the other possibility is a light amount of jaundice or else reflection from some light source you were not aware of. The other possible cause of yellow color especially on the nasal side of the eye is pinguecula, thought to come from excess UV irradiation in those who are exposed to a lot of sunlight.

Asher

Hi, Will,

Will Thompson said:

Still the color is accurate!

Click to expand...

Meaning what: that white balance color correction was applied based on a neutral target in a test scene?

Best regards,

Doug

Doug Kerr said:

Hi, Will,


Meaning what: that white balance color correction was applied based on a neutral target in a test scene?

Click to expand...

Doug,

Exactly!

However, what little tricks might Canon play in giving us the "Canon Look"?

Still, to me this is a "salmon colored image" with the skin, tube dress and wall fitting in together as if they came from the same crayon box! It's pleasing. I can't imagine that the wall and dress do not reflect light to her face, directly or indirectly.

Asher

I have no idea what the reflective color of the white of the model's eye is. If I use that as the basis for white balance color correction, I don't like the result on the whole shot. And of course the viewer won't perceive the eyes as "yellowish" in this context, surrounded by orange stuff.

Will Thompson said:

I have no idea what the reflective color of the white of the model's eye is. If I use that as the basis for white balance color correction, I don't like the result on the whole shot. And of course the viewer won't perceive the eyes as "yellowish" in this context, surrounded by orange stuff.

Click to expand...

Will, it's your photo, so if you're happy, we're all happy.

But if it were mine, I'd make some changes. Human beings do not have green eye whites. I doubt strongly that the chair is casting sufficient green to alter the color of her eyes, especially as you are using a ringflash.

Look at your photo in the red, green, and blue channels. Your red channel is almost useless. Some parts are blown out--that is, red > 255.

If it were me, I'd forget the gray card. I rarely care about gray cards anyway. Besides, you can't have it both ways. If you want to say there's a reflection causing a color bias, then a gray card is irrelevant. If you want to say you used a gray card to arrive at your color balance, then something went amiss.

Use curves and your own judgement to arrive at pleasing and believable colors. Don't use the eye dropper on eye whites for color balance. That is likely to give you weird results.

If you shoot in raw mode, you might have some luck in making corrections. If you shoot in jpg only, you will have less luck.

Do you still have your raw file?

Will,

Where's the rest of the shoot?

Hi, Will,

I'm still intrigued by this shot. The model is stunning, and the pose very fetching.

As I mentioned before, my main concern is the "washed-out" appearance.

I see that in fact in the image as posted the R channel histogram is very concentrated at the high end (there is no such thing as R>255. although of course I know what was meant).

I assume you still have the raw file from this shot. I'd like to see a redevelopment of the image with a view to keeping all channels fully participating in the overall dynamic range.

I think the chromaticity of eye whites is of no concern on its own unless it contributes to an "unnatural" appearance of the image (or of course alerts us to some serious medical problem on the part of the subject).

And of course if the color balance that seems to produce the desired result overall leaves the eye whites perceptibly off-hue, they can always be plugged (since we are not in a PJ context, where doing such might excite the sensibilities of the truth-in-photo-journalism police).

I have no concern at all with alleged violations of orthopedic surgical dogma.

I enjoy the continuing development of your skill in this genre of work. I think your models benefit from it as well.

Best regards,

Doug

Doug Kerr said:

I think the chromaticity of eye whites is of no concern on its own unless it contributes to an "unnatural" appearance of the image (or of course alerts us to some serious medical problem on the part of the subject).

Click to expand...

Green eye whites are normal or natural, Doug?

The eyes are the key to the photograph. If the eyes are out of focus or discolored, the photograph will not appear correct. The eyes are the pathway to the soul of person and photograph. To say that, "...chromaticity of eye whites is of no concern..." is poor, at best.

I have yet to read a book on retouching people that suggests that eye whites can be green. I have yet to see a cover of a magazine showing a realistic looking model with green eye whites. And those two statements apply across both genders and all ethnic groups.

On a separate topic, I wish whoever is deleting posts afterward to stop. Please leave the posts "as is."

Doug Kerr said:

(there is no such thing as R>255. although of course I know what was meant).

Click to expand...

If it can't be beyond 255, then RGB properly captures all colors? When converting from LAB to rgb, you always capture all colors because there is not such thing as R>255, B>255, or G>255? If something is blown out and RGB shows 255 in all channels, then we know that is true correct color? There are no imaginary colors, colors that go beyond gamut?

Kevin Stecyk said:

If it can't be beyond 255, then RGB properly captures all colors? When converting from LAB to rgb, you always capture all colors because there is not such thing as R>255, B>255, or G>255? If something is blown out and RGB shows 255 in all channels, then we know that is true correct color? There are no imaginary colors, colors that go beyond gamut?

Click to expand...

But how are imaginary colors, ie those beyond the RGB color space expressed? One could extend the scale beyond the 3D shape to find colors that were not remapped into that space. We just don't!

However, in any real case, where the camera does give a 255 plateau of any channel, that color is not real and a lot more colors might be missing.

The salmon color of the picture is still pleasant! That the tube dress her face and the wall are of similar hue, troubles me to a degree as I like to know how things happen, but if Will says it's accurate, then we are left just enjoying it for what it is.

Asher

Will Thompson said:

Here we go!

EF 50MM f1.2L Prime 1/60 f4.0 ISO 400 1DsMKIII MR14EX Ringlight​

Click to expand...

some very fast editing, masking and changing the background color to something more complementary to the skintone and shirt.

Hi, Kevin,

Kevin Stecyk said:

If it can't be beyond 255, then RGB properly captures all colors? When converting from LAB to rgb, you always capture all colors because there is not such thing as R>255, B>255, or G>255? If something is blown out and RGB shows 255 in all channels, then we know that is true correct color? There are no imaginary colors, colors that go beyond gamut?

Click to expand...

I said we cannot have R>255. It's true, and I didn't mean anything beyond that.

Of course, if in an RGB color space one of the coordinates is at 255, we can be pretty sure that the actual color (lets say, as represented in the raw data from the sensor chain) is beyond the RGB space gamut (and I know that's what you probably meant).

But no, in an RGB color space, we can never have R>255.

Best regards,

Doug

Doug Kerr said:

I assume this is in response to my mention that we cannot have R>255. It's true, and I didn't mean anything beyond that.

Of course, if in an RGB color space one of the coordinates is at 255, we can be pretty sure that the actual color (lets say, as represented in the raw data from the sensor chain) is beyond the RGB space gamut (and I know that's what you probably meant).

But no, in an RGB color space, we can never have R>255.

Click to expand...

Just for the benefit of others, I know Doug from other internet truck stops and have the utmost respect for his bright intellect and inquisitiveness.

I looked at this situation slightly differently. Even though my eyedropper reads 255 on the image in question, I don't believe it. Just as I don't believe a 10-volt meter maxing out at 10 volts when hooked to a 20 volt power source. I am equating the eyedropper as the meter.

If the image had been worked on in LAB, I suspect that the color might have survived. When converted to sRGB, the strong colors would be truncated.

So, yes, in RGB, you can only have 255 as the maximum value that can be shown.

I still maintain that the value is beyond 255. It is beyond what can be measured and represented in RGB. Nothing more.

Hi, Kevin,

Kevin Stecyk said:

I still maintain that the value is beyond 255. It is beyond what can be measured and represented in RGB. Nothing more.

Click to expand...

D'accord!

Best regards,

Doug

Joachim Bolte said:

some very fast editing, masking and changing the background color to something more complementary to the skintone and shirt.

Click to expand...

Joachm,

I'm pretty impressed that you proceed where others would fear to tread! You have clearly demonstrated a way of exploring new possibilities of presentation. If I had changed the color of the wall as you did, will would be after me! However, your'e probably quite safe, LOL! He's a most generous fellow and capable and inventive engineer.

Will is meticulous in his efforts and not easily swayed. However, I must admit that I, myself, find your presentation agreeable and we have lost the sense of one color infusing everything.

The use of complementary color in the wall does give her separation and adds' life and dimension.

Asher

I liked the wall orange.

Best regards,

Doug

Me too.

And I think we're missing an important point.
For all the seemingly well intended color analysis and image editing, Will stated some time ago: "This shot turned out just as intended ...".

John Angulat said:

Me too.

And I think we're missing an important point.
For all the seemingly well intended color analysis and image editing, Will stated some time ago: "This shot turned out just as intended ...".

Click to expand...

But he did provoke the debate with the title, "How bad can it look?", LOL!

Asher

True on that point!

Don't mess with Will

Will won't post much if you mess with his images.
If he wanted a green wall and white eyes, he's shoot it that way.

Kathy Rappaport said:

Will won't post much if you mess with his images.
If he wanted a green wall and white eyes, he's shoot it that way.

Click to expand...

Kathy,

Not to worry, Will is wiser! This way he gets variants he does not have to feed and house! He can claim anything his own and also disown them, the best of both worlds.

Asher

Doug:

My Eyedropper says =425, but then I think it is calibrated in ml or cm3.

Here you go, Deniable proof that the colors are correct.