by Oleg Vorobyoff
This is a revision of an article that originally appeared in Stereoscopy magazine, Number 106, Issue 2.2016, and is reproduced by permission of the author.
I presume that most of the readers of this journal have a special feeling for 3D imagery. In this article I will explore the nature of that special feeling and in what ways it can be harnessed to produce effects not found in the mainstream arts. The ability to see in three dimensions, that is to sense space directly through the two eyes, is astonishing to me. I cannot even imagine how the brain manages to isolate objects so precisely in a complex visual field despite the fact that much of what the eyes take in is visible to one eye only. The most similar art form to 3D is, of course, conventional 2D photography. It might seem that a conventional photographer should be able to transition to 3D easily using experience gained shooting 2D. But I will argue that, because a 2D print is an actual object whereas a 3D image is an illusion, it is usually a mistake to apply 2D conventions to 3D photography. Conversely, it does not make sense to employ 3D where 2D does the job as well or better. Being that the 3D illusion must be conjured using cumbersome equipment and accessories, such as dual projectors and special glasses, I feel that 3D is best reserved only for situations in which 2D fails.
3D Has Unique Strengths
What is it, then, about 3D imagery that makes it so captivating despite its encumbrances? I believe the answer lies in the degree to which a 3D image can stimulate the senses. When we look at a 2D print we initially encounter it as a piece of paper. In another moment the mind realizes the paper contains an image and proceeds to decode the image to understand its importance. I say “decode” because it is not obvious that a flat piece of paper should be anything but that. Show a dog a picture of itself and it will not react to the image at all. We have learned to understand printed matter over years of encountering documents and photographs. Presented with a document we proceed to read it; given a picture we know to interpret it as a representation of some thing. When presented with a 3D image, however, we have no need to interpret the image – the thing is right there, hovering before our very eyes. How often have you heard a person gasp upon first seeing a 3D image? A 3D image provides a direct path into one’s senses, much like music does.
I believe the key to worthwhile 3D photography is to identify which properties of the 3D medium are unique and to exploit them, either for realistic depiction of some object or scene, or for artistic expression. Since I have been shooting 3D I have been troubled by how many of the tricks had worked for my 2D art photography come up weak when applied to 3D images. After all, good art often depends on poetic ambiguities. 3D images, however, are starkly realistic, thus quite unambiguous. So I spend much of my time searching for new tricks capable of exacting poetry from this resistant medium.
A Window Versus a Frame
A print is a picture customarily bounded by a rectangular frame. A 3D view, on the other hand, is bounded by a virtual window floating in space. We tend to approach a ordinary picture much like a page in book – we read it. The picture frame powerfully confines our attention to the picture’s surface. A 3D view, on the other hand, is experienced. As we navigate the real world we constantly scan what is before us for things of interest. If an object catches one’s eyes the head turns directly towards that object the better to scrutinize it. It is just that sort of head turning moment I like to capture in my 3D views. The conventional photographer will, by habit, place the object in the camera’s viewfinder artfully offset from the center of the frame as suggested by the rule of thirds. This is likely to be a big mistake in 3D. Never compose a 3D view through a one eyed viewfinder. I always have with me a set of two-eyed viewing masks that help me find promising 3D views and to determine the optimum camera position and window size for each shot. Often an inch left or right or up or down in vantage point makes the difference between a winner and a loser.
This is one of several 3×5 inch 3D viewing masks I always carry in my shirt pocket. This particular one is cut out at an aspect ratio of 1.5 for a 35mm focal length equivalent lens.
The Center Must Hold!
The rule of thirds, of course, is the conventional trick that prevents a subject from hanging statically in the middle of a picture frame. We offset the subject within the frame in hopes of giving it some life. A 3D image, however, looks alive from the start – it needs no resuscitation. So what rule should the 3D photographer adopt to replace the rule of thirds? I like to ask myself “What am I looking at?” as I compose any 3D view. My goal is to capture that head turning moment discussed in the previous paragraph. I must confirm that in the center of the view there is at least a trace of the eye-catching feature that initially attracted my direct gaze. If there is none, I find that my 3D composition is likely to fail. The rule of thirds dictates how best to position a subject relative a picture frame. But the 3D window alone does not have the power to contain a composition like a 2D picture frame. This is because most things in a 3D view appear separated from the 3D window. They are out there floating in space. In fact, with a wide angle viewing method the window might be totally irrelevant to the viewing experience. A 3D view, not having a solid anchor at its edges, should instead be organized from the center out instead of by cropping from the edges. It is important, however, to note that the stereo pair taking parameters should complement the stereo pair viewing parameters. If the viewing angle will be relatively narrow, like for stereo cards, the window will be more prominent, so conventional compositional techniques can come into play. Nevertheless, I have lost count of how many times judges at my club competitions have said something like “The subject in this card is centered, but it seems to work.”
There is a variation of the rule of thirds that does work for certain 3D views. I call it the rule of halves. It states that the true horizon should run along the horizontal center of the composition and that there be a prominent feature or spatial transition near or along the vertical center. This type of composition can be effective because it is natural; it mimics how we see the real world as we walk about in it – head level and facing the direction of the walk.
I offset the man a touch to the left of center to define the sight line to the right as the primary. Being a seasoned 3D photographer, I was not even tempted to put him a third of the way from the edge of the view. Note how the composition expands from the center out.
This is a stark application of the rule of halves. Again I offset the bisecting corner a bit to the left of true center, but the composition would have been no less effective with the corner at dead center.
This view did not attract much interest. I believe that was because there is nothing distinctive in the center to catch the eye. I keep it in my portfolio because I like the details in the rest of the view.
Sight Lines Rule!
Another compositional trick frequently used by conventional photographers is to employ leading lines. Fine, they work in 3D as well. But it turns out that a comparable concept, that of sight lines, has become the keystone of my 3D work. To be interesting for me, a 3D composition must provide open spaces between obstacles of various kinds. Those openings are the volumes of space through which the sight lines point. Used artfully, sight volumes have emotional power — when wide they can represent a measure of freedom. Conversely, obstacles narrowing or closing off a volume of space can induce feelings of limitation, or even threat. Say there is a spiny cactus guarding the front of a stereo view. That makes passages around the cactus feel all that more inviting. In the field I spend most of my time walking around searching for a vantage point from which sight lines lay out a compelling balance with obstructive elements. But it is highly improbable the real world will spontaneously array itself into a form that speaks to my deeper emotions, so I frequently return from a photographic excursion with nothing, but how sweet it is when a view does grab me emotionally. I do not gamble but imagine this is the feeling slot machine players have on winning a jackpot. Therefore a warning: 3D photography can be addictive. See your doctor (or therapist) if you experience adverse symptoms!
Besides the sight lines visible from the photographer’s vantage point there are usually potential sight lines from other vantage points within any presented 3D field. “If I could only step around that cactus I would have a clear path to the horizon.” This makes balance between freedom and limitation in the 3D view more complicated. If there are people and animals in the view a third layer of complexity might be the sight lines available to those actors. Who is looking at whom? Is that dog about to attack? The brain is surprisingly adept at identifying such sight lines. If you have doubts that sight lines are important, try to direct your gaze at a dog sitting at right angles to you. The dog will probably blink, indicating that it has sensed that someone is looking at it. I suspect that there are neurons in the animal brain dedicated to the processing of sight lines. And remember that you and I are animals as well. I believe that the ability to trace sight lines is a fundamental component of our vision. 3D has the unique power to clarify the exact direction of sight lines in a photograph.
The narrow sight line at the right frames the photographer. One can imagine how his sight line might have looked. His shot was probably better than mine.
Photographers are taught to avoid mergers of nearby and distant subject matter because they can make a picture difficult to understand spatially. A major strength of 3D is that it isolates objects in space. The two eyes can clearly distinguish the relative positions of objects in view. This frees up a whole new class of subject matter for the 3D photographer that is unavailable to 2D photographers. But like a dieter set loose at a buffet table, a new 3D photographer is likely to binge. Everything looks so good. It is only with experience that one learns to see the difference between what in the visual world is commonplace and what is rare and worth preserving.
Sight lines point to and through volumes of space. It is negative space, there is nothing there, yet it provides much of the emotional punch of a stereo view. But let me now say a bit of the about the positive space – the actual subject matter of a stereo view. It too needs to be organized in a coherent way. But instead of organizing a subject matter graphically like a 2D photographer would, the 3D photographer must organize it spatially He or she must find a way to tie the isolated objects together. The most powerful organizing element in 3D is a plane, and the most common plane in 3D views is a floor. The floor can guide the eye smoothly from nearby to distant objects. So can an oblique wall. So can a group of objects arranged along a plane. Like sight lines, planes carry their own emotional weight. A floor can invite one into a view. A ceiling can feel mildly oppressive. A wall can close off a volume of space, intriguing or frustrating a viewer. Finally, most stereo views end in a flat wall, a backdrop, which sets the outer limits of the micro-universe that a 3D view depicts. It is helpful if each sight line reaching the backdrop terminates in something interesting or at least prominent. A sight line that goes nowhere can weaken the overall composition.
Traditionally, the function of the subject matter in a picture is to tell some sort of story, so it is best not to disappoint viewers expecting one. That story will dictate to a large extent how the subject matter in a view must be arranged. I am not a story teller, so when I am out photographing I usually look for an interesting spatial environment: a stage, if you will. If I find one, I post myself there, and wait for something to happen. Surprisingly, more often then not, “actors” will start to populate my “stage”, then a coherent story develops before my eyes, and I get my shot. I do feel, however, that 3D as a medium is not very good for story telling. Other media – video, conventional photography, radio, print – can usually tell a story more directly and more fully. As long as I am shooting 3D, the real story I am trying to tell lies within the space itself. It definitely helps if the subject matter is interesting, and the viewers of my 3D views might not be consciously aware of anything besides the nominal story, but do I hope, like the music track of a movie, 3D provides an essential emotional background for their viewing experience.
We can do compositional analysis until the cows come home but it will not mean much if the final result does not feel satisfying. Chinese philosophers aim for a balance between opposing forces, say between light and dark, as illustrated by the yin yang symbol: ☯. With 3D I believe the fundamental balance needs ultimately to be between the tangible subject matter and the space surrounding it. Find that balance and you have found the best that 3D can offer in any particular situation. To do so, it helps me to forget everything else and think strictly in terms of what I call somethingness and nothingness. My word processor objects to “somethingness” but I do want a word that describes the shape of tangible things divorced from all other properties they may possess. Happily, the word “nothingness” is acceptable to my computer. Nothingness is, of course, what is not somethingness in a given 3D view. As discussed before, sight lines are the centerlines of the volumes of space that are not somethingness in the view. Now before I get thoroughly tangled up in verbiage let me say that I do not generally find time in think in such abstractions while photographing in the field. Since digital memory is now cheap I recommend capturing as many compositional variations at a promising location as time allows. Back home it is likely that one of those shots will pop out as feeling perfectly balanced. I find that viewing the pictures as greyscale anaglyphs using Stereo Photo Maker is the best way to review a batch of 3D pictures. The lack of color makes it easier to concentrate on the purely spatial characteristics of the images.
3D Wants To Breathe
Photographers are taught to crop their pictures down to essentials. That makes sense if the purpose of a photograph is to spotlight the nominal subject. Regarding that I go back to the question I asked at the end of the first paragraph. Why not simply present the image as a 2D print? The true subject of almost all the 3D images I take now have what I would find missing in conventional photographs, mostly spatial characteristics. Now, as discussed previously, space can be organized by the size, shape and direction of sight lines. Outlining the sight lines, is obstructive subject matter. The 3D photographer must be careful not to crop out that subject matter. In the same way that the much of the power of a conventional photograph lies in its framing, the power of a 3D view lies in the integrity of its sight lines. Often that requires capturing a wider view than needed for a 2D print. I once listened to a judge who recommended that I crop out some distracting items at the edge of one of my images. I did that and was initially rather pleased with the result. Coming upon that image at a later date, I found that it felt unbalanced, and ended up restoring it to full width. Ironically, that same judge selected the original uncropped image as one of the five best of the year.
I used this cropping per the judge’s suggestion but later felt something was missing.
I reverted to this uncropped stereo pair after I realized that the stuff at the left served to define the edge of the primary sight line of the composition.
Distractions? What Distractions?
That brings up the question as to what should be considered distracting in an image. In a conventional photograph distractions usually irrelevant subject matter or extremely bright areas at the edges of an image. Such features weaken the power of the picture frame to hold the composition together. Judges recommend doing a “border patrol” to identify and eliminate them before considering the photograph finished. But remember that a 3D window is rather weak to begin with. The power of 3D to isolate an object in its own spatial niche tends to prevent any one object from dominating a composition, however near it may to the window. Since a 3D view is usually anchored by what is in its middle, that is where one should start look for potential distractions. The worst are what I call orphans and widows. Orphans are bright objects behind the outline of nearby subject matter that are visible to one eye but not the other. These can appear to float annoyingly in space in front of the view, sometimes right in front of the viewer’s nose. They can also appear along the left and right edges of the 3D window, though to less effect since the brain automatically tends to pin them at the plane of adjacent imagery. Widows are objects just partially obscured by nearby subject matter, thus fixed in space, that still manage to draw the eye away from primary subject matter. In real life we can resolve these types of ambiguities by a slight movement of the head, but a 3D view effectively locks the head in place relative to the imagery, so it is best to remove the more offensive orphans and widows. They can be removed from the stereo pairs easily in most image processing programs by a healing brush tool or equivalent.
Note the orphans behind the skateboarder’s knee and the widow underneath his right hand
The man walking behind the skateboarder could be also considered a widow, but he is not so offensive, being much less bright. Note, incidentally, how the two distant stationary figures help fix the primary sight lines to the background.
Retinal Rivalry: The Good, The Bad, And The Ugly
Ambiguity: the soul of art. And sure enough, despite its stark reality, the 3D medium does throw some crumbs to the aspiring art photographer by way of retinal rivalry, that is, incongruities between what each eye sees. It is possible to simulate movement, say of water, by taking the left and right chips of a stereo pair out of synchronization. For me such effects get old fast. I appreciate still photography for its ability to freeze time. Leave movement to video photographers. A more natural retinal effect is specular glitter. Sometimes sparkles in sunlit water or snow will appear to float randomly in space. Velvety surfaces and reflections in windows and water, too, induce retinal rivalry that can be quite beautiful and may generally left as is in a finished view.
Although this stereo pair was perfectly synchronized, natural retinal rivalry in the reflection rendered it very difficult to view. I spent a few hours in Photoshop cloning out the most painful bits of rivalry, leaving enough to retain the reflective character of the water.
The worst kind of retinal rivalry in my opinion is due to window violations. Window violations occur when an object positioned forward of the virtual window overlaps it and causes to window to appear to warp or even snap from one position in space to another. Window violations used to be totally unacceptable but nowadays it seems most stereographers are oblivious to them. It is must be said, however, that many of their 3D views are quite spectacular and win prizes. Nonetheless, I prefer to avoid violations, or at least to find treatments for forward imagery that maintain the integrity of the stereo window.
I knew the violation at the lower left would be unacceptable but the view was too beautiful to pass up. I decided at the time I took the picture that I would fix it in Photoshop.
As tiny as it is, it took me a half day of Photoshop work to tease out the rear bit of foliage at the lower left corner of the view from that in front of it.
After Impact, What?
As long as I have access to this forum I might as well voice another pet peeve. We are told that a good photograph must have impact. I think that is because most professional photography is used for advertising. You need to wake me up before I can receive your message. But an impact, by definition, is brief. So we have learned to leaf through a stack of photographs giving each a couple of seconds of time and perhaps as much ten seconds on a few that have particular impact. So that my work will stand out, I try to embed something striking in each stereo view whenever possible. However, more important to me is the resonance that follows the initial impact. I would hope that the viewer could look at my view for a minute or more. After all, I have spent hours working that photograph in Photoshop and know that there is much beauty therein to revel in. Of course, even a minute is probably too much time to ask anyone to stare at a single photograph. I wish I could attach notes to my best photographs indicating that perhaps this one is worth a minute of your time, and that one maybe even two. But that would be losing battle. I note that the slide show feature on my new TV has three settings, all of ten seconds being the longest.
A thought I forgot to include in the original Stereoscopy article is the resemblance I find between looking at a 3D view and playing on a pinball machine. For those of you not of a certain age, a pinball machine was a cabinet in which a steel ball was propelled by a spring loaded plunger into an inclined field of pins, bumpers and flippers. The player seeks to prolong the game and gain as many points as possible before the ball drops into a drain at the bottom of the incline. The plunger can be thought of as the impact of a 3D view. After the mind is done with the primary subject matter in the view, the eyes bounce around to see what else might be of interest. I believe it is important to place that secondary subject matter in a view strategically to keep the eyes and mind engaged for as long as possible.This is done by giving the eyes a variety of routes through the view, as well as pleasant rest areas at which the eyes can gain a second wind. I think I have just overstretched this metaphor and must now declare “game over”.
The main difference between 2D and 3D imagery is that 2D presents a flat graphical image imprinted on (or projected onto) an actual object, whereas 3D presents a view, a mere illusion. These two types of images are so different in nature that compositions that work in one format are often ineffective in the other. It is advisable, therefore, that 3D photographers unlearn many of the compositional habits they may have acquired doing 2D photography. Instead of concentrating on subject matter, look for sight lines. Remember that while 2D can capture what exists in the world around us, the unique strength of 3D is the ability to capture the nothingness that lies between what exists. If that is not poetic, what is?