After posting Getting the Exposure Right, I received a quite a few
questions about how I achieved the HDR (High Dynamic Range) version of
the mailbox photo (last photo in that article). This tutorial will walk
you through the basics of creating the same look by hand. All you need
is camera and photo editing software that supports layer masking (you
can follow these steps in Photoshop, Paint Shop Pro, and The GIMP, among
others).
A little background: HDR is a type of image manipulation. The goal is
to blend multiple exposures of the same scene into a single image in
order to get a result that has more dynamic range than your camera is
capable of recording. Typically this applies when you have uneven
lighting in your scene as in the example photo.

Final HDR result—this is what we’re going for.
The first step in creating an HDR image is to shoot your scene
multiple times at different exposure settings. A tripod comes in handy
here. The idea is to get at least one frame with the correct exposure
for each region of your scene. In this case, there are four regions I
want to capture full detail in: the sunset, the clouds surrounding the
sunset, the middle ground trees, and the mailboxes in the foreground. I
started shooting with the exposure set for the mailboxes and then
gradually decreased the exposure in steps of about 2/3 of a stop, taking
a shot at each step, until I got a good exposure for the sunset. I
ended up with about 10 frames but decided just to use the following four
in the final image:

Exposures chosen for HDR blending.
The next step is to load these images up in your image editor of
choice. Create a new image and stack each of these exposures in separate
layers. You can go from darkest to brightest or brightest to
darkest—whichever way you choose, keep them in order. I chose to put the
darkest layer on top and the brightest foreground layer at the bottom
of the stack.

Then I add masks to my top three layers so that they are hidden (the
contents of the masks are black). This effectively removes the top three
exposures from the image and lets the fourth, bottommost image show
through. Now grab a largish brush with a soft edge and paint white onto
the black areas of the masks where you want the associated image to show
through.
Imagine you are looking down on a set of transparencies stacked one
on top of the other. Where the mask is black, the layer beneath shows
through. Where the mask is white, the image in that layer is what you
will see. Gray portions of the mask blend the layers smoothly.
This step can be quite time consuming and you’ll find yourself going
back and forth painting different portions of the masks on different
layers to bring out more or less detail in your final photo. Start with a
large, soft brush to quickly get to a rough draft. Later, switch to
smaller brushes to make refinements.
The masks for this project ended up being quite complex. Here are the individually masked layers:

Layer 2 mask (one from bottom)

Layer 3 mask

Layer 4 mask (top)
The following images demonstrate how these masks combine together to achieve the final result:

Layer 1 (bottom) by itself

Layer 2 is added on top of that. Notice this layer was used to bring in
detail in the clouds and to darken the street for additional contrast.

Layer 3 added. More cloud and tree detail.

Layer 4 added. Sunset detail. Final result.
This technique is time-consuming and difficult to master. A good way
to ease yourself into this kind of manipulation is to try blending just
two layers together with a single mask. That is much simpler than trying
to juggle multiple masks. Two exposure masks are actually very useful
in a number of situations. In fact, you don’t even need two
separate
exposures. This same technique is often used to simply blend an
adjustment made to one layer with the layers beneath such that the
adjustment doesn’t affect the entire layer. This technique can be used
to adjust the contrast only in a portrait subject’s eyes, for example,
or to apply localized sharpening or blurring.
0 comments:
Post a Comment