People have dreamed of flying unaided for centuries. And while it’s still impossible, that doesn’t mean we can’t use photography to make it look like a reality. Making someone appear to levitate is a technique that places equal emphasis on shooting and post-processing. There’s no way we can separate the two. The end result looks like a complex montage, but in most cases levitation images are created using just two shots – one with the model reclining on a stool, ladder or another object, and a second of just the empty scene.
How to shoot it
The first shot to take should always be the one of the model. Attach the camera to a tripod and compose the shot with the model laying in the desired position. At this point you can let them relax. Set the camera to aperture-priority at the desired aperture. We recommend shooting at between f/1.8 and f/5.6, although shooting with the lens wide-open does make correct focusing and depth-of-field much more difficult. The plus side, however, is that the background blurs out of focus beautifully to make the subject really stand out.
With the camera secured to a tripod, it’s best to set ISO to 100 for optimum image quality. Ask the model to get back in position and autofocus on their face. Switch the lens to manual focus and you’re ready to begin shooting. Once the model shot has been taken, remove the stool and take a shot of the empty scene with focus locked on exactly the same spot so the two shots will match up perfectly in post-processing.
Consistency is essential
When shooting to combine two shots for a levitation image, it’s incredibly important to make sure both shots are taken in identical ways. So that means locking the camera firmly on a sturdy tripod, making sure focus is identical by switching from AF to manual focus, and that exposure is also the same for the two. Any discrepancies in these areas will make blending the two images very difficult, and in some cases, such as with focusing, completely impossible. Take your time and work methodically to ensure little, or ideally no, variance between the two shots.
Step-by-step: Blend your two images together using powerful Layer Masks
1. Copy and paste the model image
Open both images into Elements or Photoshop. Click on the model image so it’s visible in the main window, press Ctrl+A to select all, and then hold down Ctrl+C to copy. Next select the background image and press Ctrl+V to paste the model image onto it. You will now have two Layers on the Layer palette – one for the model and one for the background.
2. Create a Layer Mask and paint onto it
Make sure the model Layer is active, and then click on the Layer Mask icon on the Layers palette. It’s the rectangle with a circle in the centre. Now press D to set the colour palette to black and white. If black isn’t the foreground colour, press X. Press B to select the Brush Tool, choose a large brush and begin roughly painting over whatever they are laying on.
3. Paint around subject edges
Once you’ve painted out the majority of the supporting object, it’s time to focus on the more tricky subject edges. Zoom into the image by pressing Ctrl and +, then select a smaller brush with a medium hard edge. Now very carefully brush around the edges of the model until the supporting object completely disappears.
4. Correct masking mistakes
When masking out small intricate areas it’s very easy to make minor mistakes. Always vary brush size according to the area you’re masking out. Use the left square bracket key to make the brush smaller, and the right to make it larger. If you do make a mistake, press X to make white the foreground colour and paint over the mistake. When finished press X to switch to black.
This article was first published in the September 2014 issue of Practical Photography magazine - download back issues here.