A healthy human eye has a focal length about 20-some mm. I know this for a fact because of my wife’s eye problems being extremely myopic with a focal length of 34mm measured by an eye specialist. Now, consider a camera lens: A “normal” lens for a camera is about 49mm with fish-eye lenses having shorter focal lengths and telephoto lenses having longer focal lengths.
My question is this: Knowing all this, why doesn’t the human eye see things as a fish-eye lens does? Or, does the brain merely adjust? If the latter, how does the brain know how everything should appear to adjust from fish-eye to normal?
A normal lens for a 35 mm camera is about 50 mm, but for a 16 mm camera, it’s only 25mm.
Focal length depends on the size of the receiving field (i.e., the film or the retina). Tiny receptors have not only a smaller “normal” focal length, but also a deeper depth of focus.
being a “fisheye” lens isn’t merely a matter of relative focal length. Wikipedia gives as a definition
A “fisheye” lens really does duplicate the effect you see from looking upwards through an air/water interface, with Snell’s law bending the hemispherical world above the surface into a smaller angluilar subtense thay can be easily imaged. Part of what makes a fisheye lens what is is is the ability to perform this sort of distortion, yet keep everything in focus.
I haven’t done it, but I;'m sure I can design a “fisheye” lens meeting this definition that nevertheless has a relativel;y long focal length.
If you are talking about the field of view - i.e. why a 5mm lens on a 35mm camera has a wider field of view than a human eye - it’s because the 35mm film is larger than the retina. Focal length determines the image scale at the film/sensor, but the field of view is also determined by the physical size of the film/sensor. Same reason why a full-frame camera has a larger field of view than an APS camera with the same exact lens.
If you are asking why the human eye doesn’t have the same type of distortion as a fisheye lens, it’s mainly because what the human eye sees is your brain’s reference for what looks correct. It looks undistorted because that’s what what your brain considers to be undistorted. If something changes in your eye, your brain learns to consider that normal. My father had surgery for retinal detachment and says part of the vision was distorted for a while, but it went away. It’s not because the retina moved around to the original position, it’s because the brain corrected for it.
You don’t have to design it. Generally, if a wide-angle lens is not specifically described as “fisheye” (or “curvilinear”) it does not have this distortion. They are called “rectilinear lenses” if you need to make a distinction from fisheye = curvilinear. Here’s one comparison photo.
The brain has a remarkable ability to edit/adjust what the eye actually sees. One good example - what passes through the lens of the eye actually appears upside-down on the retina, but the brain edits it to appear right side up. People who wear corrective lenses with a large prescription or progressive lenses experience distorted vision initially but, again, the brain adjusts and after awhile the distortion isn’t noticeable (I can see if I really concentrate on doing so). The brain does a lot of processing on the raw data sent to it by the eyeballs.
I don’t think “the brain edits it to appear right side up” is really correct, because it presupposes that there’s some sort of inherent “up” or “down” in the brain’s response. It’s not like there could be such a thing as an “uncorrected” image: Such an image would have to go through exactly the same mental processing steps as a “corrected” image.
Actually the retina has only a small zone of sharp vision (the fovea) and everything that you ‘see’ sharp around of you is seen though it. With a sharp field of view of less than 2° (twice the angular size of the moon), your eye is more like a 1000mm lens. The impression of seeing a wide sharp image in front of you comes from your eyes continuously exploring the environment and your brain patching pieces of image together.
The rest of the retina is dedicated to things like motion perception.
I’m glad you posted this because it leads into my next question. How can the human eye have such a wide range of focal lengths that no camera (that I know) can duplicate with the range of any one lens?
That’s mostly just because the aperture of a human eye is much smaller than the objective lens of most (decent) cameras. But cell phone cameras also have a small aperture and wide range of focus.
The human eye doesn’t have a range of focal lengths. It has a single focal length of about 20mm. But if you account for the size of the fovea, it is equivalent to 1000mm.
There is a rather famous experiment about image inversion. Theodore Erismann built inverting goggles, and wore them for an extended period. The brain adjusted and after a while was perfectly happy with the inverted image.
The question about fish-eye versus not in human perception is probably a non-question. The eye is not rectilinear in its optics. But neither are the optical sensors in the retina laid out in any form of rectangular pattern. Our brain makes some interesting adjustments to how we perceive geometry anyway. We don’t see verticals as converging. Yet a proper rectilinear perspective should do so. Our brain sees the information it gets from the eye in the manner it wants to perceive it. This includes the geometry.
Human eyes do have a variable focal length. The eye is in effect a compound lens system where most of the optical power comes from the curvature of the cornea, but focus is adjusted by altering the curvature, and thereby the focal length, of the lens behind it. This is in contrast to conventional cameras whereby focusing is achieved by altering the distance between the lens and image plane.
