Here I present a workflow for creating a dynamic image using layers in Photoshop. Why? Well, because I like to share and because I got some requests on my Google+ album asking how it is done. To illustrate the process, I’ll use a set of images I created for Scaramanga Bags, a cool company in the UK that sells vintage leather bags and other things like journals and vintage suitcases and trunks (see the Scaramanga Concept Images here). On their website Scaramanga already has nice urban portraits with their bags, so I wanted to go in a different direction. I wanted to create portraits that convey a feeling of abstract motion. Something to invoke a feeling of movement and action. I love photography and painting. I began with photography looking for image perfection, and then moved to painting after developing a color palette in Photoshop. I like to light an image in layers, and in Photoshop I layer colors and backgrounds to add a sense of visual movement to an image. I look at a scene, put on a pair of rose-colored glasses, and I have a layered image (because at the base, this is all Photoshop does). When you can do this in your mind you then just need to translate that to something other people can see, and for that we have Photoshop. The aim of this article is to show you how to combine images together to create unique, balanced color combinations, which add a desired character to the original image.
The Basic Recipe
I generally apply this concept to portraits, where I want to add a certain character which complements the person photographed. First, begin by realizing that the person is a person, not simply a subject (A Person is not a Subject) for academic study. I start out with a base portrait image, generally shot in a studio environment with a two or three light setup using softboxes and maybe a beauty dish. Why? Because we need a decent (well exposed) portrait to start with. It should be something that speaks to you and has the look and pose you want. The layers in Photoshop are just there to modify the intention of the original image (otherwise just go ahead and create an image from scratch and render it in 3D).
I always start with a well-exposed base image that defines the main textures, tones, and colors of the person. In the Scaramanga Flight Bag images I used a Sony A900 and Elinchrom lights with a CreativeLight softbox. You don’t need an expensive camera and equipment, but you do need to know that a properly focused image with proper exposure will give you the largest amount of information to work with. If your initial image has high contrast or deep and dark shadows, then you just need to know that you can’t modify those areas of the image very much, and they will not blend so well when we layer a new image on top of it, since the very dark areas contain very little color to modify. So, let’s start from the base image.
The Base Image
In reality we’re mixing static image layers one on top of the other. In my mind I’m painting on layers of color movement to complement a portrait. I began with images produced in my apartment studio, and posed in such a way as to communicate the idea of running or of standing still, with motion in the background. This is my base, a strong pose which will be modified (enhanced) by a new layered color environment. For more info on creating a dramatic pose portrait check out my post on this subject (Urban Ninja – Dramatic Pose Tutorial). In short, I take my inspiration for poses like this from comics and graphic novels such as Conan the Barbarian, 300 and Watchmen.
After importing the images into Lightroom I chose the best and then increased the Fill Light to reduce the contrast in the image, and then exported to Photoshop for layering work. When exporting from Lightroom I don’t want deep and dark shadows, but rather a lot of information to work with and which will respond well to layering. Once in Photoshop I will often start by adding a Black and White and High Pass layers to the base image (although I didn’t need to do that for this image set). I first copy the original layer, add a High Pass filter, and set the blending on that layer to Soft Light. This has the same effect as increasing Clarity in Adobe Lightroom, but in a more controlled way. I reduce the Fill value on this layer so that everything blends well together and the image doesn’t look gaudy or like it was just run through an actions industrial meat grinder. I will often also create a Black and White adjustment layer, and then set the blending to Multiply. You can then adjust the values for reds and greens and blues. This desaturates the color while intensifying the shadows of your base image. It can darken the image a lot, but the goal here is to modify the tones of different parts of the image (such as skin tones). Again, I will often reduce the Fill of this layer so as not to totally kill the base colors.
I always start from the base portrait and then choose layers on the fly. For the Scaramanga images I wanted a lot of bright colors with movement. So, I opened up Adobe Bridge and looked for long-exposure night scenes with lots of color and light streaks. To achieve this abstract motion goal, I picked a few images that I had shot in New Orleans, Zurich and Berlin. The key here was to have images with long light streaks and pockets of intense color, which would blend in with the form of the person in the Scaramanga portraits. By blending well I mean that the lines of the night scenes would coincide with the lines of the runner (think of drawing lines over his body and comparing it to the flow of the layer images – check out my Dynamic Pose Tutorial for clarification). There’s no formula here, you just need to pick images that work well together. Aside from light streaks, these images also have very interesting pockets of color, and also recognizable object elements such as a tram or street scene, which then defines the background environment of the final image. The night images from Zurich give the feeling of running through a city of lights, while the one of Bourbon St. gives the idea of a person standing still while the environment is exploding in color around him. Now that I have chosen the layer images, I just need to blend everything together.
After picking the layer images in Adobe Bridge I opened them in Photoshop, and automatically set the blending mode to Overlay. This allowed me to preview how the different light and color elements of the layers would work together, and how the flow of the lines of the layers would mix with the base portrait. At this point, the image just looks like a couple of images stacked on top of one another, and that lazy sort of image production just doesn’t do it for me. To properly blend the images you need to play around with the blending modes, like Overlay, Softlight, etc. and also change the Fill and begin masking individual areas with a paint brush or gradients. To mask a layer by painting simply select the layer and then choose the layer mask icon. When you now paint with black, the layer will be masked (or hidden). You can change the Opacity of the brush to mask the layer gradually with each new brush stroke (the recommended method). When masking in this way I usually use a brush Opacity between 3-20 with a soft brush. This is where I act more like a painter than a photographer, masking and blending the layers uniquely together. I rarely use the entire layer image. Often I use a gradient to mask out half of it, and also paint away most of the layer over the person. I will also add full Color Fill layers (usually set to Overlay blending) to tweak the overall color. Eventually, the final image will then start to come out. To illustrate this process, you find here the secret goldmine of any Photoshop artist, the screenshots of my Layers window on my two favorite images from this set, the Runner and Bourbon St. You can clearly see how the different layers were masked, and what the original layer images looked like before blending.
If this sounds complicated don’t be deterred. Essentially all I do here is to mask out the parts of the individual layers which don’t flow well together, and in the end I have an image with all the flow and color vibrancy I desire. The main idea is that the character of the layers complements the base portrait. I save the image and open it up in Lightroom. From Lightroom I play with the colors further, adjust shadow and highlight colors, Vibrance, Clarity, etc. until the final color tones are correct and then I export.
For more info on layers and portraits, check out my Hyper-Realistic Portrait Photoshop Tutorial. This covers the main topics I addressed in this post, but you get to see a screen cast of the whole process.