Back to basics: subject isolation

_64Z5755 copy
The man: color, texture, contrast, motion. We’re not really missing shallow DOF, are we?

Judging from the correspondence and comments flying around recently, it’s about time we did a refresher course here on the fundamentals of composition and image-making. As usual, there’s far too much obsession over hardware and not enough thought about what it’s actually being used for. This will be the first of several posts from the archives in this theme. That said, those people are unlikely to read these posts anyway…

Regular readers will know that I’ve distilled down four common traits of a strong image: quality of light, clarity of subject, balance of composition and ‘the idea’. The first is very simple: does the light present the subject in a flattering way or as you would desire? Is it directional (i.e. are there shadows) so that it’s possible to determine spatial layout of the scene? The last two require some practice, and the final one is really an never-ending quest for every photographer because there is no limit to the complexity of message that can be conveyed. Today, we will look at the easiest yet most commonly overlooked one of the four: subject isolation.

At the most basic level, if your audience does not know what the image is about, then there’s no way any of the higher order associations/ implied relationships can be made. You cannot tell a story if you cannot identify the actors. The image has been limited in potential by removing or at best making ambiguous one of the main things that can make an interesting image. And if you don’t know what your image is about, then there’s simply no way for your audience to do so, either.

_7503044 copy
The two silhouettes: color, texture, contrast.

Rule number one: ask yourself what the image is about. If you don’t know, then move on. If you answer ‘it’s the whole scene’, or the ‘quality of light’, or something equally ambiguous, then one of two things must happen: if it’s the whole scene, then the whole scene must be uniformly interesting without any distractions or particular bits that stand out – that is one of the objectives of abstract photography. If it’s ‘quality of light’, then the only tangible translation of that is color, intensity and directionality – in essence, can you see the shadows, and are they framed to clearly be the subject?

All methods of subject isolation rely on one fundamental principle connected to the way the human eye/brain combination perceives the world: pattern recognition. Elements that are familiar or expected (i.e. do not break pattern) do not receive as much attention as those which are not – consider camouflage; the whole point is to make the camouflaged object blend in with its surroundings by making it appear the same. It does not break pattern and therefore gets no attention. Conversely, if you have something that does break pattern – the proverbial bright pink bull in a china shop – then it attracts far more attention than is usually warranted.

_7502760 copy
Phone woman: color, texture, contrast, DOF.

Photography is fundamentally about light. Therefore, if we go back to the visible/recordable properties of light I touched on earlier, we can also derive the mechanics of subject isolation: color, intensity, directionality. There are really only five ways* to separate one area of the frame from another: color, depth of field, contrast, texture and motion. However, the last three are really the same thing: texture is spatial frequency, which is represented by a change in contrast; motion is spatial frequency that only changes perpendicular to the direction of subject motion, which is again represented by a change in contrast. (Contrast is of course both the absolute difference and rate of change in intensity of light between adjacent spatial zones.) So ultimately, we are left with three. Of course, the more of these methods you can use, the more strongly your subject will stand out – and some won’t be sufficiently effective alone, like depth of field.

Other than if we’re working in monochrome, this is one of your strongest tools: it’s independent of overall and local scene brightness, depth of field, spatial size and frequency (texture). A bright red dot on a mostly monochromatic background will definitely catch your viewer’s attention. However, just be careful that the color of your subject isn’t overpowered by the color of secondary elements in the background – the bright red dot won’t stand out if it’s in a field of other, larger bright red, blue, green and orange dots.

_8B05694 copy
Gott’s. Color, texture, DOF, contrast – note how you notice the hard-edged menu first, and then only the person holding it – which is intentional, and a useful deployment of shallow DOF.

Depth of field
Shallow depth of field alone is the most overused method by amateurs. It is not recommended to rely solely on this method because not only does it not work to isolate a subject by itself – you can’t always use it. If you have a subject that sits in identical light to the background, with similar texture and similar color, it will never stand out no matter how shallow your depth of field. You still need contrast to make it work. On top of that, if your subject is at or close to infinity or the hyperfocal distance for your focal length and maximum aperture*, there is no way you can throw the background out of focus. However, being able to throw backgrounds and foregrounds slightly out of focus is very useful when it comes to accentuating the other methods of subject isolation whilst still retaining enough identifiability in the other elements to add context. You’ll notice that only two of the example images use depth of field as isolation techniques; and even then, the remaining elements are very clearly identifiable.

*This may actually be very close to the camera for a slow/wide lens, and also depends on your output size/resolution.

_8B11952 copy
Stream of people. Color, texture, contrast, motion.

Contrast is the only one of the five that works all the time: something that is higher contrast will always be more obvious than something of lower contrast. It can also negate spatial/size differences – low contrast and large is less noticeable than high contrast and small. Perhaps this is a holdover from our days in the jungle, when there was danger in darkness and uncertainty in sudden light. Directional light is the key to producing contrast, because contrast is really the presence of shadows – look for your shadows to be visible in some way. If the shadows fall directly behind your subject, the light source is originating from the same direction as the camera and you will have no easy means of separating foreground from background. (Though that may not be a bad thing in some cases – for instance if you are going for an abstract, painterly image – one more example of rules that are not necessarily absolutes…)

_64Z3977 copy
The rock: Texture and motion – minimal change in contrast, minimal change in color. But it works because there isn’t anything else that’s more contrasty or different in color to the rock, either.

I’ve always considered this property to be both the simplest to see and the most difficult to explain. It’s really about spatial frequency: something that has high frequency detail won’t stand out against something else that also has high frequency detail. Perhaps an example is in order: people (high frequency, irregular detail) stand out well against plain walls (low frequency, regular detail) but not so well against trees (high frequency, irregular detail).

Effective use of motion blur can take place in one of two ways – the subject, or the background. If both are moving, then it’s impossible to determine whether the motion is deliberate or accidental camera shake. Effectively, what is happening is that we are creating a patch of fairly uniform texture against which to place something else; as a result, that change in texture creates a noticeable transition and therefore isolation.

Finally, it’s important to remember that what makes your subject stand out can also make distractions stand out: your subject will therefore have to be even more well isolated to rise above the noise. It is of course also possible to have a clearly defined subject with only one or two methods of isolation if your background and edges are clean – this is the main reason I keep stressing to my students to pay attention to the immediate backgrounds. Why make life more difficult otherwise? MT


Visit the Teaching Store to up your photographic game – including workshop and Photoshop Workflow videos and the customized Email School of Photography. You can also support the site by purchasing from B&H and Amazon – thanks!

We are also on Facebook and there is a curated reader Flickr pool.

Images and content copyright Ming Thein | 2012 onwards. All rights reserved


  1. Ian Carroll posted a fascinating photo to your Flickr group long ago. I swear to God, I looked at this waterfront scene at least three times and didn’t see the subject, a ship at dock.

    Liverpool Layers

    As Ian explained, this was an example of “dazzle camouflage” used on ships in WWII. The bright colors add up to such a burst of unexpected color and disruptive patterns that the brain rejects them as not making sense in that place.

    I do miss that curated group. I don’t pay much attention to Flickr now, but do have a large collection of favorites culled from the contributors. What a bunch of talented and highly skilled photographers. I’ve often thought that I could pull together a gallery of very salable prints from it. They also serve to remind me what a hopeless putz I am as a photographer.

    • Sorry, didn’t mean to hijack your space. I thought that would show up as a text link. And while we’re at it, make that World War I.

    • Yes, it’s pretty surreal (though I thought that particular boat was an art installation). Without the masts you’d completely miss it because the silhouette is broken up completely.

      Flickr group: it was just eating up too much time; for the ~40,000 or so images in there, I must have seen close to a million. 🙂

  2. I remember this was one of the first lessons you taught me years ago, either in email school or maybe reading the original version of this article, and it’s stuck with me to this day. It probably informs a large part of how I think about composition. And it surprises me to see how many people aren’t aware of subject isolation issues, even people who should know better! Anyway, it’s good to revisit it, and something that I think about all the time when I compose (or at least try to!).

    • It usually is 🙂

      I think of it as undoing tunnel vision: it’s easy to become so focused on one single element that you forget about everything else; if the audience doesn’t have the same subject fixation (“attention span”) then they’re going to get distracted…

  3. giom grech/ Malta says:

    The two silhouettes: color, texture, contrast. Super in all ways!


  1. […] we think is interesting and hopefully arrange them in a way that makes them noticeable in the right order of prominence relative to their importance in the idea we want to convey. The overall process of making an image […]

%d bloggers like this: