Make a Photo The Palinode Way

Hey. Do you like workflow? Does the word 'workflow' sound sexy to you? Or does it sound sort of anatomical and squicky? Whichever way you lean, workflow seems to be the chosen term for photographers describing their process. Workflow captures everything from image preparation and capture to organization, post-processing and delivery to clients. Given the multitude of styles and tools available to photographers, the process is more or less unique to each person.

And hey, here's mine.

People seemed to like this photo, which showed up on this site (where no one saw it), Flickr (where a few people saw it) and Instagram (where a whole schwack of folks scrolled past it). It's an image of two women in conversation over lunch at a local Korean fusion restaurant. I make no claims about this photo's greatness, but I have a blog and my wife's out of town, so hello free time.

This may shock you: that scene presented above has been tampered with. It has been nudged about, visually massaged and, most damningly of all, Photoshopped. Let's take a look at how I went from my original shot to the one above.

Photographers tend to gravitate towards certain subject material. Some adore wildlife, which leads them off into the wilderness with foot-long lenses and camping gear. Others like natural landscapes or night skies. Some have a fascination with faces and end up doing portraits.

My primary photographic love is people: people in action or just sitting around, people striding down the street or holding hands or doing any one of the million startling and beautiful things people do. In a photograph, people become abstractions of shape and contrast, but they are shapes that form stories.

In this case I spotted two women talking over lunch, and I wanted to hold that action: one woman leaning in to listen as the other expounded. Given my position and the 35mm field of view of my lens, I knew that I wouldn't be able to isolate the subjects in the frame; there would have to be some context, mostly in the form of the man on the left. Fortunately, the rope partition perfectly set the two women off from the man but allowed me to capture three figures in the photo for an asymmetrical composition. If you can't do perfect symmetry in a photo, always go for threes or some other odd number.

When I'm out and about, I usually carry my Fujifilm X100T, which is either the greatest or the worst camera ever for street photography, depending on whom you ask. I like it because it's relatively small, relies on separate aperture and shutter controls for basic exposure controls instead of a PASM dial, and has a virtually silent leaf shutter (It's also a beautiful looking camera that assuages the disgusting hipster that dwells in the filthiest regions of my soul). I had the camera set to a 4.0 aperture and a minimum shutter speed of 1/60, which I knew would give me the right depth of field and motion capture for this scene. I've found that f4 is really the sweet spot for the X100T's lens, with just the right trade-off between depth of field and sharpness.

I took four or five shots to make sure I got what I wanted. I knew I needed the women with their faces visible, their eyes open and their features expressive but also flattering. Lots of street photographs play up the grotesque, but I like to portray people in the best possible light (unless they're jerks).

Anwyway, once I got home that evening, I imported the photo into Adobe Lightroom as a RAW file, which is essentially an 'uncooked' photo; RAW files contain the information captured on the sensor and can be thought of as the digital equivalent of a film negative (indeed, Adobe's version of the RAW format is .DNG, which stands for "digital negative").

As you can see, the RAW is reasonable but underwhelming.

The horizon is tilted, the subjects are a little washed out by contrast, the exteriors are completely blown. The first thing I did was straighten it up with the Auto upright correction tool. Sometimes this works beautifully, sometimes it's a complete shit show and you have to tweak perspective manually. Depending on distortions introduced by your lens, this can be finicky work. The 23mm 2.0 lens fixed to the camera does show distortion in certain situations, but in this case it wasn't an issue.

All straightened up! The next thing is to consider the role of colour in the photo. To wit: does it add to the story or distract from it? In this case, I didn't feel that it added anything. The photo would be a lot more compelling if the shapes and lines were emphasized - the angles of people's backs, the three horizontals and the srong verticals of the ropes dropping from the top of the frame and being picked up again in the reflection on the table.

I have a number of black and white presets that I use from various sources, but in this case I avoided presets (programmed settings) and went with the Monochrome + Yellow Filter Fujifilm camera profile. The result is a crisp and slightly contrasty look that I enjoy. Sometimes I tweak the profile controls further, but in this case I mostly liked what I saw.

Then I did what most photographers probably do but won't admit to: I went to the Basic adjustment panel and hit Auto Tone. This lets Lightroom do a series of basic exposure and tone controls. It's often a good starting point for adjusting a photo and can sometimes bring out aspects of the image you hadn't thought about beforehand.

In this case, Auto Tone did much of what I wanted: brought out some of the fine detail, toned down the contrast a bit and generally balanced the exposure to reduce the problems of dealing with backlit subjects.

After a bit of mucking around, I wasn't quite satisfied with the empty space around the bottom of the photo, so I added a -.5 exposure gradient to darken the area a bit and keep the eye on the main subject. I like to keep most of my adjustments small and subtle where possible.

I continued to make small adjustments here and there to minimize the empty space and elements around the edges of the frame (particularly the light fixtures), but eventually I decided to crop in a little bit. Cropping is a last resort for me, but if it serves the photo better, then so be it.

Perhaps you prefer the extra space along the top and right areas, but photos on the Internet are usually seen as a thumbnail first. If the composition isn't immediately strong, no one is going to pause and take a closer look.

One thing I didn't do in Lightroom is sharpen the image, which really surprised me when I was preparing this post. I apply some sharpening to every image I process, but this is one of the rare photos that doesn't benefit much from sharpening. If I were to process this for print, then I'd definitely sharpen it up. Strange things happen when an image hits paper.

There were a few more adjustments to make to the image, but they couldn't be done in Lightroom. This was a job for Photoshop.

If you look closely at the photo, you'll see some unnecessary elements: signs, lampposts and so on. It's the ugly random crap that your eye snags on. For example a parking sign just below this woman's chin:

Yeah, that's gotta go.

What the hell? Stupid content-aware fill.

Turns out the solution was more content-aware fill! That tool is the solution to, the cause of and solution to all of my problems.

I won't go through the rest of the Photoshop process in numbing detail (mostly because I'm awful at Photoshop and I don't want to show just how bad I am), but a bit of careful selection, Content-aware fill, cloning and blurring removed the signs and the lamp posts.