Start by photographing a similar situation to the one just described, a portrait in a setting, and then use a manual selection method to select just the area of the person. The tools available to you will depend on which software you normally use, but the two most usual methods would be either a kind of lasso tool, with which you can draw an outline, or mask painting with a brush.
Having done this, including refining the edge by retouching it if necessary, save the selection if your software allows this. Then make any kind of adjustment to this area that makes it stand out more clearly from the surrounding, while still looking realistic. For example, you might use a curves adjustment to increase the contrast, or alter the colour balance.
For this exercise I decided to use a recent image I had taken of our neighbours cat as although the bottom half of the shot is quiet tonal, the top half is predominantly white, which should allow me to complete the task at hand.
So that I could make the necessary changes to both the background and subject of the photo, I created a new layer, which included just the cat. I made this layer by using the lasso tool, set to magnetic as I found this an easier option for capturing the information I needed, although due to the setting of this tool I had to decrease my pixel count to 24bit RGB, which in turn gave me less RAW information to play with.
I now had two layers to play with, the first being the background with no cat
And the second the cat with no background
The first thing I did was to conduct a crop to tighten things a little around the cat. As I wanted an even tone to play with when changing the background information, I set about the unenviable task of cloning in the detail missing with the deletion of the subject. There is probably a much easier way of doing this, (I did try the fill tool, but the results were less than perfect); so due to my lack of both experience and time I cloned in the missing pixel detail.
Once complete, you can see from the layer information included, that I conducted quiet a bit of manipulation to get what I thought was a better representation of my garden, and I am really pleased with the detail contained in the fence panel. Once I was happy here, I moved onto the cat and felt that implementing a small s-curve not only lightened him a little, but also brought out more detail of his fur, although I could not manage to keep the detail in his eyes and pink ears.
You can see from the final image produced, things are far from perfect, as although the image works okay, you can see the extent of the manipulation.
The most obvious is that the cat looks placed and this is due to the lack of shadow underneath his resting body. Although there is quiet a lot of information contained in his fur, due to sloppy use of the lasso tool, some of the detail is lost against the background and although when standing alone each component looks right, when merged together they do not really work.
What you have just done is the direct equivalent of traditional dodging/burning under a darkroom enlarger. Few people in the days of film considered this anything but legitimate processing, and yet … The selection is in a sense arbitrary, based not on the objective qualities of the image but on the subjective photographer’s eye. The effect, of course, depends on how extreme an adjustment you make to that selection, and what kind of adjustment. Consider the limits that you would accept for this to remain an innocent, legitimate adjustment. Reflect on this issue in your learning log.
I think, that by conducting this exercise, my feelings towards image manipulation have strengthened, and I am against it even more now than before.
As we move through these exercises, I am beginning to understand, that with knowledge, patients and will, anything is possible, and as stated in the coursework above, random choices made by the photographer can contribute to huge changes within images. I can see where the quote ‘shoot in RAW, then change your image to however you want it to be’ has come from, those lazy photographers who would rather spend hours in front of their computer over getting out and taking real photographs …
I think that there are times when this kind of manipulation is acceptable; especially when you can see that the results are in stark contrast to how a scene would ordinarily look, such us placing an object in a scene where it does not belong, or changing the background detail to something out of sync with the point of interest. Nevertheless, it is those times, where it is difficult to see the seams and joins that the subjective lines become blurred and the limits of manipulation are reached.
I can now see why academia was so set against digital manipulation 20 years ago (see my post ‘Digital Photography and Truth’). Okay, so, the world embraced manipulation back in the days of film, but it took much more effort on the photographer’s part to achieve than within a digital darkroom, where only a few well-organised keystrokes can produce the same effects and more. I am both saddened and annoyed by those who have no qualms about making such big changes to their images. We have seen that by moving pixel information from one place to another, images that are more desirable can be produced, but by moving and duplicating pixels, we are changing the DNA of our images, and making them something they were never meant to be.