31
\$\begingroup\$

Occasionally I'll find myself as the photographer for a professional event at my Alma mater. As a natural consequence of the demographic, there's quite a few people there with severe acne and also scarring. I want to take flattering pictures of these people since they put in a lot of hard work into these events and they deserve to be recognized, however it's something that makes people very self-conscious and delicate to broach.

Photoshopping it out (referring to more than just a few zits or pocks) is a possibility but events are geared more to volumes of photos than they are single highly polished ones like portraits, and this makes it not worthwhile to sit down and heavily repair dozens of photos, especially for the compensation. What can I do using my current equipment and workflows to make the most flattering pictures?

My equipment:

  • Camera: D3200 with 35mm f1.8 lens, hotshoe flash and diffuser head -- usually I step back and open the aperture as wide as I can get away with to soften features and shrink flaws

  • Post: Bridge and Camera Raw -- I'll add additional softening and de-luminate/desaturate the reds. Reduce contrast. Desaturate or even go black and white.

Photography takes place anywhere from brightly day-lit offices to dim soirées. Sometimes I get lucky and I can pose people so it's minimized.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ The content aware heal bush can make manual fixes faster, but really, I don't know that there is a whole lot that can be done other than offering makeup before the photo. \$\endgroup\$
    – AJ Henderson
    Mar 26, 2014 at 16:39
  • 1
    \$\begingroup\$ could you post some example? \$\endgroup\$ Mar 26, 2014 at 16:47
  • 7
    \$\begingroup\$ @woliveirajr not sure someone would like to stumble across their picture in a help forum, on a question about how to remove their skin problem.... \$\endgroup\$
    – db9dreamer
    Mar 26, 2014 at 16:58
  • 1
    \$\begingroup\$ @dav1dsm1th :) I was thinking more of a partial image, like just above-mouth-below-eyes, or a forehead. You're completely right, no one would like to have a photo in a "how to I make this person less ugly". I should have asked more specifically... \$\endgroup\$ Mar 26, 2014 at 18:00

3 Answers 3

25
\$\begingroup\$

Usually, people are aware that they have it, but don't want it to be so pronounced.

If you can use Gimp, there is a plugin called Wavelet decompose.

In this site there is some information on how to use it to retouch pictures.

I like that technique because you don't need to complete remove what doesn't please you, but you can minimize it so the picture still looks natural but without all that impact of some acne or so on.

Edit

I don't know how your models look like, so I took one example from wikipedia that, by it's turn, took it from some US government site.

What the plugin does: Basically it separates the image in many layers, and each layer contains some information from the original image. Each layer has some detail level, allowing you to edit each layer/level independently of the others. For example, color will be on one layer, small spots will be on another.

Below, I think you can guess which one was the treated image. I used that plugin to decompose the image in 6 layers, and edited just 2 of them (painting what I wanted to change with neutral 127,127,127 gray.

Treated photo

It was a very fast job. It took less time than the following picture, that anyone can see was done very lazily.

Decomposed

Here you can see the original picture (1), and it decomposed in 6 frequencies. The (2) is the residual image, where you have mostly the colors. (3) to (8) are the decomposition, with increasing frequency (so in (3) you have less detail and in (8) you have the highest - finer - detail).

You can see that (4) and (5) have great variation of color, so editing them reduced the acne/scars. The post-processed image still have acne and so on, but very lighter, still making the model look a real person with skin imperfections but not so strong.

\$\endgroup\$
4
  • \$\begingroup\$ This is a much more complicated version of ordinary frequency separation, which can also be done in Gimp without installing additional plugins. Seven layers to edit separately is not simpler than two. It also requires installing and becoming familiar with Gimp, which is not exactly the answer somebody who has Photoshop would be looking for. \$\endgroup\$
    – user2719
    Mar 27, 2014 at 16:55
  • \$\begingroup\$ @StanRogers are you sure he has Photoshop? He just says Bridge and Camera Raw, which make me think of Adobe Bridge and Adobe Camera Raw, not Adobe Photoshop. If Photoshop had a similar feature, I wouldn't turn to Gimp to do that specific kind of edition (Portraiture has almost the same result, but don't have the same degree of control over the result). \$\endgroup\$ Mar 27, 2014 at 17:16
  • \$\begingroup\$ I have both PS and Gimp. \$\endgroup\$
    – BB ON
    Mar 27, 2014 at 17:53
  • \$\begingroup\$ You can have it in Photoshop using Wow! Tonal Equalizer: knowhowtransfer.com/apps-photoshop-extension-filter-plugin/… \$\endgroup\$
    – Royi
    Feb 16, 2015 at 9:26
14
\$\begingroup\$

To be completely frank up front, there's a reason why there's a market for professional (or at least expensive) Photoshop plugins, and you've just found it. High volume with quick turn-around at low production cost is the market driver. Spending a couple of hundred bucks on a plugin to save hundreds and thousands of hours of processing down the road is more than worth it. But it's hardly what anyone wants to hear if they're working on a tight budget.

To do the job as quickly and effectively as possible, the images that need the most work are going to have to go to Photoshop. (Or to an image editor at least, but you seem to have Photoshop handy.) The Gimp answer already posted mentioned wavelet decomposition, and that's pretty much the sort of thing that has to happen to make things easy: frequency separation. Separate the colours/tones from the details. There are a lot of tutorials out there for frequency separations in Photoshop, searching on YouTube will give you hundreds of them that are all more-or-less the same.

Most of the colour mess (the redness of current problems and the "lumpiness" of deep scarring) will live on the low-frequency layer, and can be attacked in an aggressive and almost sloppy manner. Not that you want sloppy, of course, but the point is that it's not a finicky process. (This part can be almost completely automated using a plugin at its most ridiculously overdone setting.) You may find that most of the things you thought were problems simply vanish into the "normal" range. Anything left can easily be healed/cloned on the high-frequency layer without worrying too much about the colour. (Again, a plugin can help, but this time set to "stealth mode" where it does very little.)

Even in a fully manual version of the process, you can get an awful lot done in very little time. You're not going for the fashion/beauty look, just for something that makes people look like they're having a particularly good day, and that can be something like two to five minutes per processed image (less with a fast computer and a little practice). A plugin like Imagenomic Portraiture, an action in Photoshop and a fast workstation can make it literally a one-minute-per-image deal with great results.

\$\endgroup\$
0
1
\$\begingroup\$

Besides post-processing, a soft focus or diffusion filter will soften detail. With proper lighting it may also add a halo effect to the subject.

Example images of diffusion filter versus none

http://www.bhphotovideo.com/c/product/457719-REG/Tiffen_W52DDFX2_52mm_Digital_Diffusion_F_X.html

\$\endgroup\$
3
  • 2
    \$\begingroup\$ This image is all wrong. It shows two different photos, not the same photo before and after. Look at the mode's eyes, the glasses and her gaze. If you recommend a product that does something, please show before and after pictures of the SAME motif, not different exposures. \$\endgroup\$
    – teylyn
    Mar 27, 2014 at 7:03
  • 7
    \$\begingroup\$ @teylyn The product is a physical filter that takes time to screw on a lens, and the nature of demonstrated feature requires a human model (preferably alive one). The pose and lighting in images looks similar enough to draw conclusions; unfortunately, the low resolution hides any effect of the filter. \$\endgroup\$
    – Imre
    Mar 27, 2014 at 11:47
  • \$\begingroup\$ Agreed not the best picture. \$\endgroup\$
    – kmarsh
    Mar 31, 2014 at 16:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.