WPPC (Weekly Post-Processing Challenge) - Oct. 12, 2014

Hi William,
That is a link to your portfolio, we cannot copy the image from there. You need to insert the larger version in your post here...

 

I was able to copy the image from William's portfolio (control-click on my Mac laptop's trackpad... is that right-clicking for others?).
=========
(I worked on this entirely apart from any consideration of the final portrait.)

An interesting image. It didn't appeal to me at first because of its low-key nature--I couldn't discern the subject well enough--, but I took that as the challenge!

The first thing I wanted to do was to increase the exposure (+2.95 in photoshop) to see if I could recover the details in the reflection, especially the eyes, since her back is toward us. This worked pretty well: it revealed the intensity of her stare at herself and a degree of some kind of emotion (fear?) in her large, lovely eyes. The downcast gaze of the woman behind her, visible only in the mirror, and another(?) figure off to the picture's left, also only in the mirror, added a bit of mystery. But I did not like the man (<waving hi at William?>) visible directly behind her reflection.

I also wasn't happy with the resulting color. It seemed too harsh. The blue of her T-shirt was too intense, and the same with the green wire. The brick, too, seemed unpleasant. These drew away from the secondary mirror image.

I tried desaturating... a lot (-84, I think--all this in Photoshop CS6). That got me somewhere.

But I was still thinking it needed something. It was too... crisp. The blue was still too bright. All that seemed at odds with the "grungy" setting. Now, sometimes I like that, and I thought about just toning down the blue, but then I got to thinking about how we were seeing the scene.

The idea that it might be viewed through another pane of glass--that we are looking uninvited onto a scene of private tension--came to me.

So, wanting to tone down the colors even further but not quite make it monochrone and to add a bit of "distance," I decided to play around with opalotype settings in Topaz Labs BW Effects 2 (I first undid my previous desaturation). I dealt with the green wire after, with the hue/saturation sliders, and reduced the saturation overall by about 22% (I think), because there was a bit too much yellow.

I couldn't successfully deal with the blown-out highlight of the pipe arising from her head, so I cloned it out. I don't usually do that in my own images, but since I was doing that anyway, I also cloned out the white wire. (If I'd had the raw file, I'd have worked harder at recovering detail from the pipe and wire.)

The opalotype filter ultimately resulted in darkening the man's head into abstraction (sorry, William!), adding a sort of fogginess around the edges, as if viewed through a smeary pane of glass that hasn't been fully wiped clean, and gives a certain vintage look to the image, as if time has faded the image... but not the intensity of her stare.

 

Like Noreen this image did not initially appeal to me either. However, it is a challenge and I certainly need to get more comfortable with post processing so here is my attempt.
The original image seemed way to dark and I wanted to see more of what was reflected in the mirror. After importing to LR 5.6 I bumped the exposure up to +2.30. My goodness I can now see several people in the mirror's reflection that I did not see in original image.
My image is severely cropped because I wanted to eliminate the ugly wire at top left and the black spot on the girl's t-shirt. This also allowed me to focus more on her t-shirt logo (Be Brave & Shave!) as well as see more of her eyes in the mirror. Wish I could clone out pipes coming from her head, but I have no clue or talent to do that. I also adjusted the luminance to -54 to make her shirt darker and intensify the lettering.

 

First increased exposure then cropped all but barber, young woman in mirror and the back of the young woman. I left part of her arm and the drying rack showing intentionally - seemed to lend a more personal aspect to the woman and where she was.
Next I desaturated - the colors seemed too bright for the apparently odd circumstances of the photo - and burned in the background. I also dodged the face of the youn woman in the mirror as well as the back of her head as an afterthought to give it kind of a glowing appearance.
Ultimately the slogan "Be brave and shave" seems to contrast with the dark feeling of the photo as well as the seeming uncertainty on the young womans face int he mirror.
All editing done in CS4

 

Liked B&W

 

I removed the pipes and wiring with the cloning tool and cropped to feature the main subject. I adjusted the lighting in Color Efex. I then created a black and white version and blended 80% luminosity to the background copy. My version was mostly inspired by the way she looked at herself...

 

William, thanks for a fun challenge. It allowed me to do some experimenting with layers. My choice, which I did prior to looking at anyone else's to avoid any potential influence, was to modify the image so that the reality became blurred and the reflection became the focus, symbolizing how the subject or any of us might reflect on what life might be like under less fortunate circumstances.
My procedure was to first crop the image and remove some of the more distracting extraneous items via the clone tool. Then I made a duplicate layer and applied Gaussian blur, erasing the blur over the mirror. I decreased the opacity somewhat to allow the writing on the t-shirt to be legible. Then I applied an adjustment layer and darkened the image, again erasing the mirror area. I applied a third layer and adjusted brightness to bring out the mirror image a bit more, this time erasing the area that was not over the mirror to maintain the darkness there. After merging all the layers, I reduced the contrast some and applied a little sharpening.

 

My immediate reaction was that I did not like it. Unesthetic and much too busy. A snapshot that tried to catch a glance in a mirror and a message on the back of the woman with lots of things being transmitted to the viewer around. For me, the event is mainly to be found in the eyes of the mirrored image, so why not concentrate on that ?
So, I cropped it drastically, corrected the verticals (ruler and image rotation tools); used Neat Image to treat the noise; passed it through the Curves tool to lighten it up; and finally used Unsharp mask - all in PS CS5. Finally I changed the title into: "Be Brave and Save"

 

Hi all, my first entry into these challenges.
Like many here, my initial reaction to this photo was mild, to say the least. Turning it into an arresting, or even interesting image seemed futile, however that also presented itself as a challenge.
This is the end result of a bit of lateral thinking. You guys be the judge, but I think it shows why one should never give up on or dismiss ones own images out of hand.
For the record, although quite a dramatic shift from the original, this was a pretty quick edit...once I'd decided on the direction I wanted to take it...around 10 mins actual work. Just goes to show that some of the simplest techniques can have the greatest impact.
Interested in hearing your feedback.
http://www.photo.net/photo/17881675&size=lg

 

Indeed, "challenge" is most appropriate in dealing with this week's WPPC. I also found the original very busy and overly dark, especially the mirror reflections.
Initially I did some adjustments in PSE, starting with slightly straightening the image and cropping to eliminate primarily the faucet and drainboard. I then converted the image to B&W in NIK Silver EFEX Pro 2 and used that software also to increase detail. Finally I used Color EFEX Pro 4 to add luminosity so that the subject and the refections are both clearly visible in the frame.

 

Bill Jordan. Brilliant interpretation.

 

Joel Clarke. What did you actually do to the photo, as in what steps did you take using what software? We are here to learn what processes we can all use in post-processing not just to see the final result. Thanks for the outstanding interpretation!

 

Thanks bro! Only you would use my name and 'brilliant' in the same paragraph.

 

There are obviously not only many variants of post-processing, which have been introduced, like in the previous two weeks, up till now.
What may be new here is that we seem also to have very different interpretations of the initial image of William, which influences the post-processings.
My own take on it is, that I think the real strong visual element, apart from the shaved head, is the two questioning/challenging eyes of the girl in the mirror. These eyes, as I see it, tells the whole story. Her own felt bravery and maybe also the challenge to the other girl in the mirror with her hair intact, but maybe the next in line for the challenge. What I did was to concentrate on these two and cutting out the rest, but adding as title the main message: Be brave and shave (add I have already done above).

 

Anders, I agree that the reflection in the mirror is the key element in the shot, which is why I also focused primarily on that in my edit, but I also think you lose a bit of context if you don't include the slogan on the t-shirt. In retrospect, if I were to process it again, I might crop a bit more off the bottom as others have done, but still keep both the girl and her reflection. But I do like that, in your version, the other girl, with hair in tact, also becomes a key element as someone to be challenged or merely a point of comparison with the main subject.
And as my brother questioned above, I'd be curious to know Joel's reasoning for the inclusion of the red overlay. I think the interpretation was similar to mine, but obviously portrayed much differently.
For those who chose to go with a B&W version, I'm curious to know what the thinking was behind that (other than Lupo, who describes why he did it based on his interpretation of the subject).
I find it interesting to see how much interpretation actually goes into post processing beyond simply trying to make the shot look pretty (though there is some merit to that for some subjects as well).

 

"you lose a bit of context if you don't include the slogan on the t-shirt"

Bill I fully agree, the slogan is part of the context needed to understand what is going on. I moved it the title. For me the formulation of a title or the decision not to use titles, is part of post-processing of a photo.

 

I like this photo. It's the type of behind the scenes candid snap I wish I'd taken, and often do take. And tough to edit while retaining the essence - the feel of a gritty, candid snapshot. I wanted to emphasis the youthful strength and curiosity in her hunched-forward position. The photo reminded me of being in boot camp and seeing dozens of guys with exactly the same look, staring at their reflections with freshly shaved skulls.
While I was tempted to clone out the stuff on the wall, this sort of thing can present ethical challenges. Back in the day, it was considered normal even in photojournalism to retouch out that sort of thing. But nowadays it's taboo in most journalism settings. Since this is potentially a PJ/documentary type photo and might have to pass muster - and a publisher might request the raw or original unretouched file for provenance - I left that stuff intact.
Mostly I fiddled with global tweaks in the b&w red/orange channel mixer to brighten the faces in the mirror, and to dim the surroundings (dragging down the yellow). One of the nifty things to artificial indoor light is that it tends to redden faces and skin, while casting a yellow tint on inanimate surroundings. So in monochrome conversions you can brighten faces using the orange and red sliders, while darkening distracting backgrounds by dragging down the yellow. If you have access to the raw file, using the "cloudy" or "shade" white balance setting can resolve some tricky color-to-monochrome conversions too, especially in photos of people with a range of skin tones from light to dark.
Not much selective brush retouching was needed on the reflected faces - just a barely perceptible hint of brushwork. Oddly, the color-to-monochrome conversion along with chroma noise reduction did most of the work in clarifying the eyes of the gal with the shaved head. Just one of those quirks of digital editing - sometimes the global tweaks like the channel mixer, noise reduction and camera calibration/white balance tools solve specific small problems without needing to resort to brush retouching.
The rest is just cropping, to get her eye line up to the upper third quadrant (damn you, rule of thirds!). And a tiny bit of vignetting using the paint overlay mode in Lightroom, which helps tame contrast without too-obvious oval vignetting darkening. And I added a bit of faux-film grain because the original JPEG seemed too smooth, and it seemed to suit the gritty surroundings.

 

I like these threads! Fantastic idea! Lots of positive karma -- opportunities to learn and to hone skills!
So I did a color and then a monochrome conversion -- all CS5.
For the color, I first corrected the horizon. Then I created two layers for the two lighting planes. Everything reflected in the mirror was far more distant from the flash and therefore had to be normalized with the foreground. After normalizing the reflection, I merged layers and applied a new curve to fine-tune the overall exposure.
Then I created another "dodging" layer to lighten areas of the reflected face. I hate CS5's dodging tool, as it makes saturation go nuclear if overapplied. (PaintShop Pro has some nicer tools, including lighten, darken, color/hue/saturation to target, etc., only they were still 8-bit tools, last I checked.) I adjusted my curves and dropped saturation in the "dodging" layer and then used a mask to carefully "dodge" in the eyes. That expression is so important! After dodging it the way I wanted, I again merged layers.
Finally I used the clone tool and did some burning to clean up some of the heavy flash reflections off her head in the foreground. And of course I removed the plumbing coming from the top of her head. I kept the tonality a little bit on the low key side, overall. I adjusted the black point to make the figures in the background more murky. The hard lighting is a fact of the image, so I tried to keep the tones a bit harsh and gritty. I also wanted to draw more attention to the reflection than the foreground, and I think my curves, burning and dodging did that.
Finally, I did a monochrome conversion, because I thought the colors detracted a bit from the impact of the image. I used the channel mixer tool to emphasize the reds and darken the blues. I also blew out the highlights just a bit more than with the color image. I really can't decide whether I prefer the B&W or the color, but here they both are:

 

Oops! Too large... sorry...
Here it is again:

 

Here's the monochrome, not quite so low key:

 

Sarah,
Thanks for the detailed explanation! I generally prefer color over B and W for no reason other than my aesthetic preference, but in your case, I'm finding it very difficult to choose. For whatever reason, I keep coming back to the B and W as the preferred version.
On anothre note, is there really much difference in working in 16 bit vice 8 bit? My version of PSE is older and 8 bit, and I'm wondering if upgrading will have much of an impact (assuming the later versions of PSE work with 16 bit images).
Thanks,
Bill

 

Thanks, Bill! I've always loved B&W, because it's so forgiving of over-the-top adjustments. I think if I had desaturated my color image a bit more that it would be a bit better.
I don't think there's a huge difference between 8 and 16 bit editing. For most purposes, they're probably about the same. However, when you're doing a lot of detail work on an image, the 16 bit will come out cleaner. Of all the tools I've used, I think the "smudge" and "blur" tools shows the greatest difference between 8 and 16 bit. The problem is that of cumulative error, and you can see some of the problem reflected in the histogram with "combing."

 

Bill / Mark, my apologies for the delayed response to your questions. I should have included my workflow and reasonings along with my initial post. Better late than never though and thanks to non-destructive workflow I've been able to quickly re-create the steps I took.
Firstly, before I start any work on an image I ask myself these questions What's the end purpose of the image? What's it trying to say? What does it need to say? How is communicating this to the viewer and what if anything does it need to do this better?. This sets the framework for everything that comes afterwards. I don't make a final decision of what I want it to look like at this stage, just the message.
For this image, the message is obviously the "Be Brave and Shave" slogan but at first glance this wasn't an image that was communicating anything brave. There were also the issues of the distracting elements of the fittings on both the bricks and tiles, not to mention how the joins in the mirror tiles cut right through the subjects face.
Inspecting all the elements though, several details began to emerge that could help communicate the "brave" message by creating a somewhat scary environment. Half-hidden faces in the background for-instance - especially the one showing just a nose and eye to the (viewers) right of subject, the security camera in the mirror top-right and those eyes.
Decision made, and referencing the look of a D-grade slasher movie / film-noir (Bill re the red overlay - partly for the slasher movie look and partly to attract viewer attention as I'm assuming this image would be used in advertising) I've done the following; all in Photoshop CC.

Open image
Levels adjustment > set highlights and shadows in each channel using sliders while pressing the alt (command) key. (I find this the quickest way to correct any issues with colour balance and will always start with this step.)
New blank layer > select layer > clone stamp and patch tool to remove the wall fittings
Crop image to tighten composition and remove the distracting sink fittings.
Double-click on background to convert it to a layer > Group layers > convert to smart object
New gradient adjustment layer > Green to Red overlay at 100% opacity > change blending mode to Vivid Light > select gradient mask > brush tool > black brush > opacity 15% > paint on the mask to burn in the subject's reflection in the mirror and selectively change the lighting on the rear wall by revealing the original image (creating both tension and diagonal lines that draw the viewers eye towards the subject in the process)
New blank layer > select layer > clone stamp and healing brush > fix three distracting areas highlighted by the previous step 1. shadow across chest 2. top lip 3. to the right of her thumb
New blank layer > fill 50% grey > change blending mode to overlay > paint on the grey layer with 10% opacity brush with white to subtly burn in the subjects eyes, security camera, faces, mirror joints
Still on grey layer > change paint brush colour to black > opacity 10% >paint on layer to create a vignette around the frame and improve the contrast on the brickwork.
Group all layers > create smart object > duplicate smart object twice
Select layer copy 1 > run high-pass filter set at 1.4 > apply an inverted layer mask > paint on layer mask using a soft, white brush set at 10% to selectively sharpen areas in the image, text, etc. building up the effect slowly if needed by repeated strokes
Select layer copy 2 > apply gaussian blur fliter at 2.7 > apply an inverted layer mask > paint on layer mask using a soft, white brush set at 10% to selectively heal any remaining destractions (these where mainly on the head and due more to the quality of the JPEG file than the editing process. I doubt they would have been there from a RAW base image.
Save as > JPEG >
Done!
This whole process took less than 10 mins time, longer than it took to write this. The process is completely non-destructive and also contains the editing history as any point in the chain can be jumped into at any time....even a far future time.
Whether or not it conveys the intent as defined in the first step, well, I'll leave that up to you to decide.