Dissection 011
An in-depth deconstruction of processing
Today I propose an extensive dissection. Think of this not so much as a glimpse into the way I process every image I shoot, but as an exercise, a way to visualize how an image can be deconstructed into multiple parts. And that’s the underlying principle: although I’ll rarely go these extremes, the basic questions always remain—what’s strong? What’s weak? What must we see?
You may notice I’m not using a previously published image; it’s not even a “good” picture. In fact, I’ve deliberately chosen an outtake:
X100F, ISO 200 1/640s f/2
Shot in JPEG and almost exposed for the highlights, but not quite: shadow detail is basically gone and there are still irretrievably blown highlights in the frame. Jacob was rolling a yoga mat after gym class (1) and I captured this very quickly, resulting in a slightly soft focus on Jacob due to motion blur.
STEP 1 - Investigation+Style
Our first step is to understand the amount of data there is to work with. Since the main problem in this image is underexposure, let’s hit either the Brightness or Shadow sliders (remember this is a JPEG file so raising Exposure will quickly prove much too aggressive on those hot highlights). Here’s Brightness set at maximum:
And here’s Shadow:
I set these to max to get a sense of how far I can go. If we zoom in on Jacob, we can see how much detail is recoverable: his hands, the yoga mat and the folds in his clothes look very 3-dimensional to my eye.
Because the light WAS 3-dimensional. Knowing this, we can get to work. Let’s reset and start over. I’ll apply RUBICON on its own layer and adjusting for brightness and shadows (based on what I just saw), using a mix of sliders until I reach a middle-ground. The mantra here: I’m NOT trying to fix everything in one go, on one layer. The goal is to view the image as individual zones. This is what I get after tweaking:
It already makes one heck of a difference, doesn’t it?
STEP 2 - Radial Vignette
Those highlights are still bugging me, but the room is looking pretty good overall. I can get to those later. What I want now is to bring my subject to the forefront, without destroying the mood that’s beginning to appear. So instead of pushing, I’m going to pull: I add a Radial Gradient Mask over the subject area—Jacob and the yoga mat—invert the mask and create a custom vignette by adding Contrast and lowering Exposure:
Again, this is mild. We’re slowly building up. But it’s just enough to begin inverting the dynamics towards Jacob as opposed to the room. Why not push the “Jacob/mat zone” instead? Because the contrast there is fragile and I prefer doing this in a more subtle way.
STEP 3 - Luma Sculpting
Next I turn to Jacob himself. I need him to pop more and to achieve this, I again draw a Radial Gradient Mask and apply a Luma curve that will slightly raise both shadows and brightness:
But I don’t stop there. I also use this same mask to address the motion blur: I add Structure and a bit of Sharpening to tighten up the edges. To compensate for the slight noise these settings cause (it’s a JPEG file, remember), I also add Noise Reduction.
If you look at the before and after, you can see how much this one layer affects the impact of the image.
STEP 4 - Highlights
Facing facts here: this is going to be a compromise because those highlights really are blown. On a JPEG file, no amount of fiddling is bringing them back, but we can minimize how much they pull at our eye. Using the brush, I draw a rough mask over those areas and then adjust until they’re no longer quite as distracting.
Using the Luma Range function, I restrict this to highlights. 2.
STEP 5 - Dog Toys
No, those aren’t Jacob’s toys ;)
But I do want them to stand out more. This is an easy fix: Linear Gradient Mask from the lower-left corner. Done.
STEPS 6-7-8
At this point we’re adding seasoning:
6. Jacob looks a little faded, so I add another curve on him to adjust contrast. 3
7. There’s a wall on the right-hand side of the image that should be dark: it isn’t because of all the adjustments we’ve made. A quick Linear Gradient Mask fixes this. 4
8. One last global exposure tweak. I often do this on a new filled layer to avoid messing around with previous settings.
FINAL TAKEAWAYS
So, here’s the before and after:
The background has no adjustments at all except for a bit of vignetting. And here’s the final layers list:
I never name my layers btw, so this was all for your benefit ;)
Having said that, this would be an awfully good habit to get into.
I call your attention to the opacity percentages in this list. The Opacity slider on each layer is incredibly useful because it essentially allows us to tweak multiple parameters at once: by lowering Opacity, we reduce the intensity of every adjustment we’ve added to it. As you can see, I use this feature quite a bit. It’s the ultimate seasoning function.
As I said at the beginning of this article, this was an exercise—I rarely go this far to “bring an image back”, so to speak. But while it may seem daunting, what’s important to keep in mind is this modular nature of our images. It doesn’t mean we should treat every single photo as a space mission, but knowing what’s possible and being able to quickly pinpoint strong and weak points can open up avenues we may not have been aware of.
Over time, this becomes second nature. And the cool thing is: we can’t break anything in the process.
...............
Remote classroom, obviously.
I’ve dropped the brightness slider which could affect mid-tones and shadows as well.
I could’ve used Luma but I default to RGB...the settings were to mild to affect colour.
I used the Refine Mask function to add feathering.
