Compose With Your Feet!
November 18, 2012
Too many of my students fail to extract the best possible images from a location because they don't compose with their feet. After laboring uphill to our shooting location on the shore of Dream Lake, they wipe their brows, extend the legs of their tripods until their cameras are at eye-level, plant their feet, and remain rooted to the spot throughout their shoot. They do zoom in and out and point their cameras left, right, up, and down, seeking the best possible composition, which is smart but not sufficient. To really maximize the graphic potential of the scene, you have to move the camera as well and look at all possible compositions.
For starters, consider the impact on your composition of placing the camera at different heights. A broad field of wildflowers provides the perfect example. Let's say you're composing a grand landscape, with flowers in the foreground and distant peaks in the background. For any given field of flowers, there is some ideal height of the camera above the ground. Position the camera at eye level and the field will seem expansive, occupying a large fraction of the frame, but sparse, since a high camera position makes the foliage in between the petals look large in proportion to the petals themselves, as seen in Figure 1. (Figure 1 is actually my new masterpiece. I call it Powerade Bottles in Bear Creek Park. Hey, cut me some slack ‒ it was almost Thanksgiving and nothing had been blooming in Colorado for two months!)
Position the camera at ground level, so all the flowers line up one behind the other, and the field will look dense but shallow, occupying only a small fraction of the frame, as seen in Figure 2.
The ideal camera height will provide just the right balance of appealing density, making the flowers look lush, and breadth, making the field look expansive, as shown in Figure 3. This ideal height will vary, depending on the density and breadth of the flower field.
Once you've considered all possible camera heights, try moving closer or farther away from your foreground subject. Remember that perspective, meaning the relative size of foreground and background elements, is determined solely by camera position. The focal length of your lens has no effect on the relative size of foreground and background objects. Don't believe me? Try this experiment: shoot a scene with a telephoto, then, without moving the camera, shoot the same scene with a wide-angle. Crop the wide-angle shot down to the same angle of view as the telephoto. The two shots will be identical, both in the size relationships and in the depth of field (a topic for another day), as shown in Figures 4-7, below.
Stitched panoramas at first appear to violate the camera-position-determines-perspective rule. Take a look at the two shots below. I shot the image in Figure 8 with a 16mm lens. I shot the image in Figure 9 using a panoramic head and a 24mm lens, vertically oriented, making three overlapping frames and stitching them together. Notice that the 16mm shot has the same angle of view as the panorama horizontally and the same 2:3 aspect ratio, yet a wider angle of view vertically. Notice, too, that in the stitched panorama both foreground and background elements are larger. The perspective, however, remains nearly the same. The bottles have the same size in relation to the background mountains in both images.
Let's look at one more comparison. If I crop the single-frame 16mm shot to have the same angle of view both horizontally and vertically as the stitched panorama, it takes on a roughly a 1:2 aspect ratio, as shown in Figure 10, below.
It looks as if the 16mm lens stretched the scene horizontally by about 25 percent. How can this be? In the first version of this post I speculated that the answer lay in the additional distortion inherent in the design of a 16mm lens compared to a 24mm lens. I was wrong. The answer lies in my choice of projection in my panorama-stitching software. The image in Figure 9 shows the result of using a cylindrical projection--the image you would see if you were standing at the center of a cylinder with the image wrapped around the inside of the cylinder wall. When I re-did the stitching using a rectilinear projection, the stitched 24mm shot matched the single 16mm shot in horizontal and vertical angle of view as well as aspect ratio.
So which to choose, single frame or stitched panorama with the same angle of view horizontally? It depends on how fast things are moving. Can you really shoot three frames quickly enough to capture the light and keep clouds and flowers stationary? If you have the time and gear to shoot stitched panoramas, go for that solution: you'll get a much bigger file, which means much better resolution in a big print.
Whatever focal length or capture method you choose (single frame or stitched), the real key to finding the best composition is to move. You'll often read advice to "work the scene," meaning shoot a variety of compositions. It's good advice if the light isn't changing, but you can't "work the scene" during the fleeting moments of peak sunrise and sunset light. Instead, you should look at all the possibilities in advance, dial in the best composition, then nail it as the light peaks.
One final tip: I have often found that people seem to appreciate photos more if it looks like moving the camera even an inch would significantly change the composition. The photo looks like a singular record of a unique place rather than an image that could have been shot anywhere within a hundred-yard radius. Add obviously fleeting light to the mix, and the photo stands out even more. You can achieve this type of composition most easily by getting low and close with a wide-angle lens, then moving the camera forwards and backwards, left and right until you have placed the foreground elements precisely where you want them in relation to the background. Maybe I should have called this blog post Compose With Your Knees and Elbows!