14

I saw this question but feel that it doesn't fully explain how this works.

I am completely blind, so I am trying to come up with a tactile pictorial representation of how field of view works.

So far, I've imagined that vision is like a funnel or a cone. If you peer through the hole, the shape of it would allow you to peer up, down, left right, and diagonally in all directions. You could make the view narrower or wider by using smaller or bigger funnels.

I've also read that the reason lens are curved is to allow the light to bend in different directions, seemingly to create an impression on it, which is to make the image appear more clearly.

Now, when people say that a camera has a 120-degree view, are they sayingthat you can see 120 degrees from your nose to your left, right, up, and down, as well as diagonally? Or, are they saying that you can see 60 degrees in all directions, and 60*2=120?

I am curious to know about this because I am thinking of getting some Horizon Glasses for a visual interpretation service called Aira, which claims that their glasses have a wider view than the smart device's built-in camera.

  • 1
    Angles are inherently 2-dimensional objects. So field of view described as angle measurements also span only two dimensions, which are typically horizontal, vertical, or diagonal. When it matters, the plane of interest is specified – horizontal field of view. If it doesn't matter, such as when comparing "the" FOV of two lenses, the plane can be left out. (A 35mm lens will have a wider FOV than a 50mm lens. This is true regardless of whether it is the horizontal, vertical, or diagonal that's compared.) – xiota Jan 11 at 5:04
  • As far as the glasses are concerned, you'd probably better understand what's being described if there are different models for you to try. Have you considered learning echolocation? – xiota Jan 11 at 5:05
  • 1
    I cannot utilise echolocation very well because I have significant hearing loss in both ears, but not severe enough to be considered deaf. That clarifies some things, though, so thanks. If you can make that into an answer, that would be even better. – HeavenlyHarmony Jan 11 at 5:12
  • I don't think echolocation is relevant here — this is an assistive device which delivers images of the world to a sighted human assistant. – mattdm Jan 11 at 17:46
  • 2
    @xiota I think an angle is 1-dimensional, its a scalar value. A single angle can also be within multiple dimensions, i.e., the angle between two vectors in cartesian space relates to 3 dimensions. azimuth and elevation can also be spherical coordinates. when using the angle from a persons view, a way to think of it is a mirror in the center and rotating the line to the left or right, up and down. – user-2147482637 Jan 12 at 14:38
16

When a field of view is described as 120°, that refers to the total angle. So, 60° to the left of center and 60° to the right.

Most camera lenses show a very restricted subset of the field of view perceived by the human eye and vision system.

It is probably the case that the system is measured across the diagonal from corner-to-corner of a rectangle, because that's the longest and therefore most impressive-sounding number. But for this purpose it doesn't really matter if it's the diagonal or horizontal field of view — both are close enough.

To get a tactile feel of what this means: put both of your arms straight out in front of you. Now, widen them so each is about two-thirds of the way to straight out from your body — basically, they're the legs of a triangle with a 120° angle. This camera will approximately record anything in that cone.

I don't know what your experience of sight is and I won't presume, so forgive me if the following is obvious to you. If it is not, I hope it will be helpful.

Human vision is not primarily accomplished in the eyes — there is a huge amount of processing in the subconscious brain. This means that most people are only dimly aware of the concept of a field of view. Vision is only sharp in the very, very center, but the eye moves around and the brain builds up a mental model.

If I consciously just look forward, I have vague awareness of things off to my side. I've read that this extends to 170°, but because of this mental model I'm pretty convinced that I can perceive things directly sideways from me and even a little behind. This is my brain lying to me, but it's a generous lie because in most cases I move my head and eyes around enough to continuously refresh that area in my subconscious model.

Meanwhile, looking straight ahead, I have the general perception of about a 90° cone that I'm "looking at". Maybe a little less than that... somewhere between 60° and 90°. Even though the sharp part of my eye only covers about two degrees, I have the perception that this whole area is sharp and has my attention.

In photography, we call a lens that has a horizontal field of view of about 45° a "normal lens". That's because if I make a photograph with such a lens and print it at a standard size of around 10" (25cm) and then hold that print at arm's length, the perspective shown in the picture will more or less fit naturally into my direct view of the world.

If I took a picture with a lens that gives a 120° view and printed it the same size, it would appear to have distortion, because I'd be bringing things normally at the periphery into what, on the print, is my straight-ahead vision. So, even though the camera system you are looking at has a field of view narrower than the human vision system's peripheral view, it's definitely in the range that we'd consider "wide angle" in photography — in fact, "ultra wide angle".

| improve this answer | |
2

The lens field is circular, and does have an angular cone of view, like your funnel. However, the camera sensor captures a rectangular area within that cone, typically which just tightly fits into that cone. Meaning, the diagonal of that rectangle just fits the cone (normally, but there can be exceptions with smaller sensors). The cone has an circular angular field in degrees, perhaps some case fits your 120 degrees, which is 60 degrees up or down or any direction from center. But the rectangular camera sensor cannot see all of that circular area, so the actual rectangle size is also a limit on field of view of of the camera.

The lens field of view is sometimes referenced as this diagonal, as the maximum span it can see. Or often the rectangular width and height in degrees is specified, matching the area the the rectangle that the camera sensor can see.

The corners of the rectangular frame has the most optical problems, like aberrations or vignetting, because the corners are the furthest area from the central lens axis.

Perhaps one useful way to perceive camera field is that zooming the lens to 2x focal length reduces the horizontal and vertical field dimensions to half (1/4 area). Or 4x zoom reduces field to 1/4 dimensions, etc.
Not speaking of a half angle, as the angle is not linear (the tangent in trigonometry is not linear), but the field WxH dimensions are linear with focal length.
Or if putting the same lens on a camera with a sensor of half dimensions limits the cameras field of view to half dimensions (1/4 area).

My site at https://www.scantips.com/lights/fieldofview.html has a Field of View calculator that when given the sensor size and focal length, it computes the Horizontal, Vertical and Diagonal field of view angle, and also the field dimensions at any specified distance.

| improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.