Better still, photographers can take complete control over their multiple exposures by making them in Photoshop. This way, one can choose the ideal subject as well as the ideal secondary image—or images—and position them precisely as they see fit. It’s a technique that works especially well with portraits. Here’s how.
Starting with a studio portrait in order to make use of its clean background, I convert the image to black and white using Lightroom. One might argue it’s better to send the full-color image to Photoshop and use a black and white adjustment layer in order to maintain maximum image-forming data in the Photoshop file, but for simplicity’s sake, I like to make the black-and-white conversion in Lightroom.
You can always continue to make tweaks to the grayscale—boosting the contrast, for instance, as I did—once the image is open in Photoshop. I did this in order to make the shadows and dark tones even darker because it’s in these areas that our secondary image will show up.
With the original image now open in Photoshop, and in black and white and looking just how I want it, I then choose the secondary image for the composite. Of course, any image might work, but I like images with strong patterns and textures. Natural elements—flowers, foliage, trees, landscapes—tend to work quite well, as do cityscapes and other images with clear, identifiable shapes that will read well even when some of their image-forming information is removed. But of course, please try to experiment with any image—from another portrait to a macro image to, well, anything you like—in order to see what happy accidents you may find. That is, after all, how we break new ground!
For the portrait example here, I selected an image of some foliage comprised of different tones and textures that I thought would be bold in the composite. I think the image works well. I converted it to black and white, opened it and copied and pasted it as a new layer on top of the existing portrait image. It was a bit bigger, so I used the transform tool to resize it down to just barely bigger than the portrait frame—that way there would be no hard edges (from where the secondary image stops) sneaking into the frame.
Next, in the Layers Palette, set the layer mode on the secondary image (the top layer) to Lighten. This will tell Photoshop to look at each pixel on this layer and the portrait image below and display the lighter of the two pixels. In practice, it means that you’ll be adding the secondary image in particular to the shadowed areas of the original portrait. It’s in this way that an image with larger areas of shadow—perhaps even an image that uses split lighting to put one side of the subject’s face in darkness—work well for this technique. I repositioned the secondary image simply by using the hand tool to click and drag until I was happy with the details and texture of the image as it appeared in the darkest areas of the portrait. At this point, the image looked interesting, with the subject’s dark gray shirt largely replaced by foliage and the face, hair and background largely intact and unblemished.
Most of the secondary image details don’t, at this point, interfere with the subject’s face—but a few do. In order to eliminate them, simply click to add a layer mask and use a paintbrush with a hard edge to paint away the details where you don’t want them. I use a hard edge because a soft paintbrush edge will create a sort of feathered, blurry, more recognizable edge when a harder, crisper edge is likely to blend in better based on the image. I left some of the foliage details where they encroached on the face, but in places that seemed inoffensive. There’s no hard and fast rule here, but in this way, the edit looks more nuanced and the image more cohesive than if the pattern is everywhere but the face. This same approach will come in handy after the next steps as well.
Because of the primary image I selected, there remained lots of light areas that looked a little too plain to me. But fear not, you can take the same approach but this time make the darker details appear in the light areas. To achieve that, copy the secondary image layer and change the mode from Lighten to Darken. Now you’ll see all those light areas in the background, face and hair filled with the background image. It’s a bit much, obviously, so I did several things to bring it under control.
First, I inverted the layer to make it a negative image. What was black is now white, white is now black, and so on. In this way, the subtle highlight details from the previous version of the foliage image are now subtle dark details, and they show up nice and clearly against all the bright parts of this portrait. They showed up a little too well, actually, so I dialed back the layer opacity to make them a little less dramatic and distracting. Then I used a layer mask to paint away these details and textures from my portrait subject’s face and hair—painting away more of them from the face and leaving some of them in the hair. I then dialed back the paintbrush opacity to 50% and again painted on the mask to take down some of the remaining details that caught my eye as too strong. Then I duplicated this layer and flipped it both horizontally and vertically, then repositioned the layer to where the texture was complementary and looked good. I repeated the layer-painting process to eliminate distracting details again, and then finally I duplicated the layer one more time and repeated the rotating and masking process to add the texture from the foliage more densely across the background. Since there are no rules, it’s done when it looks right. This looks right to me, and I like how the multiple exposure effect can make a plain portrait look pretty interesting.