TECHNICAL DETAILS
Leigh Making a Watercolor of a Sweet Potato, 2020
Sony A7iii
Sony Zeiss 16-35 f/4 G OSS
ISO 100, 16mm, f/11 at various exposures, blended
Natural light
EASTER MORNING LIGHT
We are home bound, COVID’ed into place in Centerville amid the pandemic, boxed in and needing creative outlet. It was Easter Sunday morning. Leigh had departed bed early as usual, and I lingered. Minnesota in early April can be a cruel place. Sunny and springlike, but ready to turn on you like a cat you’ve pet the wrong way: from purring to claws you before you know it. And this Sunday, the claws were out. It was snowing steadily in the morning, where the previous afternoon I had been washing things in the driveway in a t-shirt. I needed coffee, and I needed to do something creative.
As I came down the stairs, Leigh was at the kitchen island, watercolor accoutrements splayed across the counter, her focus on a large and passed-its-prime sweet potato that had been lurking in a basket next to our toaster for some time. Sprouting, but still firm and edible, there had been several time across the last week we had remarked how it needed to be consumed. She looked serene there with her tools, palette and the vegetable. Light was pouring in from the window at the top of the stairs behind her, and I wondered if I could turn the scene into a compelling composite portrait with just the ambient light, no strobe.
As I thought of how to relay this effort in an overdue blog post, it struck me that I’ve never really done a full process description, or a before/after image series of one of my composite portraits. For any that saw the development of that selection reach a fuller expression through the Creative Spaces project, the behind the scenes view gives a more detailed look into the effort required to put one together. I thought this would be a good example.
I set the camera low at the edge of the counter on my largest heavy tripod, hovering right over the paper towel she uses to clean the brush, with all its mottled colors. I wanted to look across at her and create a sense of depth leading to the window at the top of the stairs. The sweet potato, palette and the group of watercolor pencils could all be used as elements to lead the eye toward her sketch book and her face. When I shoot these with an exaggerated foreground, the widest 16mm setting of the Zeiss 16-35mm f/4 lens is ideal, providing controllable distortion, wide field of view, and the opportunity for great depth of field within each frame. 16mm is not a typical focal length to use with a portrait, and placement of the subject within the frame is critical to getting a normal looking profile without too much exaggeration from the extreme width of the lens.
Normally, for a portrait like this I would shoot with the Sony camera tethered to my MacBook Pro and would make extensive use of composition and exposure tools in the software to control the camera. But I’m also generally using strobes and light modifiers. For this image, I was using only available light, and Leigh was mildly protesting that I not take too long while she finished the watercolor. I skipped the tethering and shot directly with the camera.
Even with the great depth of field afforded by shooting at 16mm, this type of shot requires a focus stack. At 16mm and f/11, with the front element of the lens only a few centimeters from the sweet potato and the paint-stained paper towel, you’ve only got an inch or two of focus depth for the foreground elements. I metered exposure for the scene, and then with the camera on full manual and manual focus enabled also, I ended up making seven exposures for the focus stack to capture foreground to infinity. I had Leigh hold quite still, but not critically so, for the stacking images. Then I made a few exposures of her with focus directly on her eyes to get the expression and posture I wanted for compositing her into the focus-stacked scene.
Here’s how the image used for Leigh looked straight out of the RAW file. Note the the foreground is completely out of focus.
The next step is to apply lens corrections, base adjustments and color grading to achieve the tonal range and color values I want for the final image. I know from experience that the later processes applied as I produce the image tend to darken it substantially, so I tend to start with a fairly bright and vibrant presentation. So far, everything is handled in Adobe Lightroom. After the base adjustments are applied, I also review the color palette, and in this case I selected a LUT (Look Up Table) to map the color values. For this composition, I selected the Hilutite LUT from Lutify.com. The foreground is still not stacked, but the corrected base image of Leigh now looks like this.
Once the tonality and color grading is completed, I can use Lightroom’s synchronization tool to apply the same adjustments and LUT to all the images for the focus stack, so they also match the portrait shot. The resulting stack is shown in the following image, but I will not use that pose of Leigh for the final. You can see that the hand she’s holding the brush in is in a different position than the final, but now the foreground is sharp all the way to the farthest elements in the image. Focus stacking is done in Adobe Photoshop by bringing the seven images into a single workspace, aligning them to ensure they are perfectly overlapped, and using the software’s autoblend tool to stack the images and produce a single, flattened output layer. It usually does a decent job, but sometimes a few manual adjustments are needed to correct stacked areas that the algorithm had out of focus.
Now the fun begins! All subsequent steps are also done in Photoshop. I’m done with Lightroom except for maintaining the catalog. For compositing the image of Leigh I chose into the stacked background, I prefer to work with an inverted mask. I placed the image of her on top of the stacked layer, added an inverted (black) mask and painted on the mask with white to reveal the composite I wanted. Then, I generated a flattened layer and began the enhancement process, the first of which is exposure blending. I duplicated the flattened composite and added an exposure adjustment layer to darken the copy. In this case, I darkened it about -3 stops. Then I blended the two exposures to bring out the drama in the lighting to what I had previsualized in conceiving the image. I again used the inverted mask technique, but this time with the brighter normal image on top. In blending, I also varied the opacity and flow of the round, soft brush I paint with to achieve the desired look. Here is the resulting exposure blend mask, the lighter areas on the mask producing the brighter resulting parts of the final image:
And here is how the working image looks like after exposure blending:
Next for some dodging and burning. I added a 50% gray layer in soft light mode and used white and black brushes for the dodging and burning, respectively. I varied brush flow, opacity and size frequently when painting. My goal here was to emphasize certain elements, de-emphasize other areas, and bring an overall sense of drama to the lighting. Here’s the result:
Getting close! At this point I could really start to see my previsualized image appearing on the screen. Now for some extra punch. I like to use the Orton Effect selectively, as its glow and blur also helps highlight certain elements while allowing others to visually recede. The Orton Effect is a three-layered group consisting of a flattened base layer, a second layer applied to the first with screen layer mode, and a third applied layer with a moderate Gaussian blur added and set to multiply mode. The opacities of each of the two applied layers are varied to achieve the right balance of effect. Then, to the top blur layer, I add a normal mask and brushed black to hold back areas of critical sharpness from the blur. The overall effect is one of intensification and it brings a significant element of drama to the image. Here it is with the Orton layers added:
The final touch is a set of layers to make final dodge and burn adjustments from the Orton group, and lastly some selective sharpening. For sharpening, I usually apply a high pass filter to a flattened working image, and set that as the top-most layer in overlay mode. I use an inverted mask, and paint in only selected areas that target the viewers perception of sharpness. These are generally around eyes, glasses, the hair framing the face, select edges of clothing around the upper body, the hands, and in this case, the edges of the sketchbook. And this brought me to the final image, which you can see in larger size at the top of the post.
Overall, this image was straight-forward and not as complex as some of my composite portraits. It ended with stacking and compositing eight separate exposures, and the resulting PSD file in Photoshop has 14 layers with masks, correction layers, dodge and burn, Orton Effect and sharpening groups. I hope you’ve gained a bit of insight into the process of creating a composite portrait in this style. My overall editing time (not including the shooting) was about four or five hours. Here’s a look at the final layer arrangement in Photoshop.
Leave a reply