How do quirks of the human vision system affect photography?

Question

When looking out at the world through our eyes, it's easy to take for granted that our perceptions are an absolute rendering of How Things Are. But in fact, it's not quite as simple as that.

Most of our actual vision is done in the brain — the eye (while amazing for a biological construct) is really a rather mediocre optical device, but all of its quirks are processed out into a smooth, high-resolution, three-dimensional model of the world. An infinite number of wavelengths of light are resolved into the perception of specific colors. Lines and edges are processed specially. Faces and other special patterns jump out at us even when they're just suggested by an alignment of shapes.

How, in a nutshell, does all of this work? And more importantly, what knowledge of this system is useful in composing photographs? What can we take advantage of, and what quirks present problems to be worked around?

Answer 1

There are several key points, of which I will pick my top ones.

1. The human vision system will re-focus very quickly, and only at what it is looking at in that moment. It is therefore difficult to look at a scene and see any kind of out of focus blur. This will cause humans to not realize the effect of certain areas being out of focus — you can't trust your eyes, unless you train very carefully for it.
2. The human vision system quickly adjusts for contrast, making it difficult to see just how contrasty a particular scene is. Thus you can't trust the un-trained eye to get the right shot. It takes practice and skill to get the right dynamic range translated.
3. We see in 3 dimensions, which can sometimes be difficult to translate into 2.
4. Even though we have a uniform perceived field, actually the resolution in the central part of our vision is much higher than the edges. Also, the edge part of our vision is better at dark perception than the central part. Also, we tend to ignore things that aren't our key point of interest. What all of this translates to, when we are taking a picture: we don't see the trash can in the scene, or the pole sticking out of someone's head, unless we specifically look for them.
5. We don't actually create a snapshot at any one time — we operate something like a movie camera, but each "pixel" operates independently. Still, our mind is able to make it look like we have a snapshot. It's therefore kind of hard for us to really understand shutter speed.

Answer 2

You're very wrong about the human eye being a very mediocre optical device.

The brain processes the image, but your eye gathers and focuses the light, for being so tiny, and made out of flesh, it's quite amazing. So amazing, that scientists still haven't been able to replace it.

The real quirk is to train your mind to think like the camera, and not the other way around.

The camera see's everything. So do your eyes. The difference, your brain filters out "noise" as meaningless information. It records it, but at a higher level it's discarded.

It's why two people can look at a scene, and when asked to describe what happened, will give two very different responses. Or why, when you go to review your photograph, you notice the powerline that was there, that you missed when taking the picture.

Your brain is great at filling in the information games, and great with filtering out information so you don't get overwhelmed. As you start to develop a "cameras" eye, you'll start seeing things as they actually appear, and not as you think they do.

Answer 3

Actually, there is an aspect of the subjectiveness of human vision that I would like to highlight. I said subjectiveness because what a camera see (either digital or analog) is what goes on the photo-sensitive sensor/film, and the information is just what colour hit that surface.

In human beings, what actually sees is the brain, rather than eyes. Human brain performs lots of adjustments and interpolation to the graphics (let me use this word) that come from eyes.

All quirks in eye vision are fixed by the brain

Blind spot

In 2-eyes vision, each eye covers the blind spot of the other eye. In 1-eye vision (finger experiment) the brain will interpolate the visible part to form the invisible part of the blind spot, so that's why you ultimately won't see your nail but still see a finger

Focus

To quote Pearsonartphoto, human eyes focus (and converge in stereoscopic vision) on what is being looked by the brain. I believe that the speed in focusing is helped by 3D vision and shape recognition - yes, I'm talking as a computer science expert, which is what I am

Centre-resolution

Again, quoting Pearsonartphoto. The central part of our retinas is higher in density rather than peripherical zones, and is almost the only part being able to distinguish (actually, colours are worst distinguished the increasing distance from centre). So if you wanted to mimic human vision in a photograph, you would have to blur all what's not in the middle of image, but then you assume the viewer will look only at the centre (or to a specific object).

I can clearly see what I'm typing in this line, but I can't distinguish the words 3 lines above this one, nor its very beginning!!!

A note about 3D vision: one mistake done by some 3D movie directors is to add focus to the 3D scene, actually forcing the viewer to look at the character that is speaking, or doing something important. Sometimes, I would like to look at the fancy dress of the guy behind him. What has this to do with photography? Simply that a photograph currently gives you freedom to look at what you want in the scene. Emulating quirks of human vision (like unfocusing a background object/character) is realistic until you discover that you deprived the viewer from the freedom to look

Shape recognition, noise reduction, vectorial storage (wow, I'm starting to be a computer paranoid!)

I think that the most important quirk in human vision is that we don't store images in memory as for they appear, but for what they are. The opposite of this is enhanced in people with photo-memory. Try to go on Facebook and pick up a random girl's photo (well... if you complain par condicio let's say a boy's photo, as soon as he's haired enough) and look at her hair. Then search for another random girl with the same hair colour, without reading further. Try to memorize the two faces. The day after, come back here and answer my question: who has the darkest hair?. When you see the two girls separately, your brain memorizes their hair colour as a stereotype, so you'll remember to have seen two blondes, or two brunettes, etc. Even if you still remember them, you won't probably be able to perfectly compare their hair colour, while you could if you had their photos at hand and read my specific hair question. If you write a program that compares hair colours in photos, you will get always the correct result.

The colour example was an example. What I said applies with objects, etc.: a photo stores the pixel matrix of colours that compose an object, your brain stores the object along with its attributes (a black and empty pen, a half-full bottle of milk...). When you "recall to mind" these objects, they are actually rendered in your mind like if I was rendering a computer-generated scene of a video game. I just tried the experiment: I quickly looked at my desk, with my keyboard, my game pads, etc, and tried a few seconds later to "recall it to mind": I forgot to render my 3D glasses and my phone, and I imagined the game pads in a different position rather their real one, and also thought that the right game pad had the keyboard attached (it was the left).

A photograph, on the contrary, in an undeletable snapshot of a continuous (analog) or discrete (digital) light signal. These signals are always affected by noise, that is removed by our brain. Look at an old videotape with some artifacts and try to remember the movie the day after: as soon as you don't focus on the artifacts (because I told you so) you'll probably remember the plain scene!.

Anyway, this question if fortunately not about computer vision, otherwise I would have started writing a book ;)

I hope my answer is comprehensive.

Answer 4

Regarding the latter part of your question, how our human vision and its "quirks" affect how we use a camera and how we photograph the world around us. I think the single most valuable piece of advice I have found regarding photography was from an article on Luminous Landscape, part of the series "Aesthetics and Photography" by Alain Briot titled:

"How to See Photographically"

What I am talking about here is seeing photographically, or seeing like a camera. How to learn how to see like a camera is the purpose of this essay.

...

Creating photographs is all about seeing and in this sense it is no different from other two dimensional arts such as painting and drawing. Creating photographs is really about studying and practicing "the art of seeing".

The art of seeing, while it may seem mediocre and obvious, is not as easy as it sounds. As mentioned by the other answers here, the human eye and the brainpower behind our "vision" is a subjective engine, constantly and automatically working to make sure we see most clearly what we need and want to see, and pushing all the other "noise" in a scene to the background.

When we look at a beautiful mountain vista with our own eyes, we see only the key things we want to see, or what we are drawn to see. The mechanics of our vision take care of "eliminating" the distracting and useless elements, and impose a vision of astounding beauty. The moment we take a photograph of the same scene and chimp it on site, were immediately confused at how boring and commonplace it looks.

A simple analogy I've come up with since reading Alain's article is this:

The art of seeing is the seeing of art.

It is not enough to see a beautiful scene with our minds eye, and capture it the same way. You must also "see" with the camera's eye, and purposefully, explicitly compose the scene you originally envisioned with your minds eye. The latter is where artistic vision comes into play. This is the process of taking direct control of our vision center, seeing everything, and subtracting until our "camera eye" and "minds eye" match.

I think a key distinction between the technological eye in our hands and the biological eye in our heads is that the former operates under a fixed set of algorithms, while the latter is adaptive and under the control of conscious thought. We can change aperture, focal length, and sensitivity with our camera, but we can literally change the vision of our mind to suit our needs. All thats needed to achieve that is training and practice.

---

A few of the key factors in seeing photographically, the factors that help you change modes from "human sight" to "camera sight", include the following:

1. Seeing
2. Abstracting
3. Focusing
4. Composing

These aspects can be followed explicitly, in order as a sequence of steps, as a form of training. Approaching each scene this way can help you develop your own mode of "camera vision" to complement your normal vision, and bring a whole new level of seeing photographically to your work.

Seeing

Before you can work a scene, you must first see it. Not seeing like a human, but seeing like a camera. Looking at a scene, and reversing the automatic simplification your mind does to help you abstract a scene and focus on the important parts. Expand your vision and observe everything that is.

Abstracting

Once you see everything before you, you need to begin reconstructing what your minds eye saw originally. You need to abstract your expansive vision, separate the different elements, identify what is necessary, what isn't, what is primary, what is secondary. At the same time, recognize the non-visual elements of what you are experiencing...your sense of emotion, the physical feel of the scene, the sounds that are present. Even though they are not visual elements, they are factors of what you are "seeing", elements of emotion that can still be portrayed via artistic license as you compose and eventually process the final image.

Focusing

Now that you have identified the elements of the scene before you, its time to contract, to zero in on the important aspects...to focus your vision on what matters and what draws your attention. Localize the abstracted elements of the scene that you wish to keep, and those that you wish to discard: What is interesting, and what isn't? What's important, and what is irrelevant "noise"? What can be done to interweave some sense of emotion into the scene?

Composing

Now that you know which elements of a scene you intend to capture, you can finally begin to compose. Composition is framing the focus of your scene while discarding the rest. Composition is involving depth, perspective, and placement of key elements according to natural rules such as thirds or the golden ratio. Composition is the technological equivalent of everything your brain did automatically, in an instant, when your eye first caught a glimpse of that beautiful thing your now trying to photograph. You're now seeing photographically, artistically seeing the art before you.

The "eye" may be simple in the grand scheme of optical systems, it may be small, it may be rather mediocre by our considerably advanced technological standards...but it is still a truly, uncompromisingly amazing device that sees, abstracts, focuses, and composes a thousand times a second. It will let you make the decisions when told, see what you want to see. It is the best tutor your artistic side could ever ask for.