43
\$\begingroup\$

This is another perspective distortion question. I understand that perspective distortion is primarily a function of capture and viewing distance, but I am still working on fully understanding the causes of the "flattening" of scenes that is credited to telephoto lenses.

This seems to refer to our ability to discern the differences in size attributable to differences in distance from the lens for subjects beyond a certain distance from the lens. If this is correct, then is it just that with the longer distance from the subject when using a telephoto lens, this compression happens closer to the foreground, rather than in the mid- or background with a wide lens?

Is it fair to say that the compression effect occurs at a certain distance from the photographer regardless of lens, but the field of view influences how close to the foreground this effect takes place? Can you share any information to help people like me understand this?

Improve this question
\$\endgroup\$
3

3 Answers 3

Reset to default
29
\$\begingroup\$

The flattening or compression effect is not caused by a particular kind of lens, it applies to all lens in the same way. Actually, this property of lenses applies to our own eyes as well. The factor that affects flattening is the distance from the camera to the subjects.

Consider the following exercise:

Place two friends 1 meter away from each other. Place yourself in line with them, so that you are 1 meter away from one of them, and 2 meters away from the other. Move a bit to the side, so that you can see both. From this position, you can easily estimate what the distance between your two friends is. Now walk 100 meters back along the imaginary line determined by your two friends. Again, move a bit to the side so that you can see both. Can you really tell now how far away from each other they are? Will you see your friends in a significantly different way if one of them moves an additional meter apart from the other? You won't, because from 100 meters away, a difference of a meter is not significant.

The thing is, our eyes have a fixed field of view, so in addition to flattening, we experience a scale reduction as we move farther away from the subjects. With a camera you can magnify the effect by using a long lens. But the compression will be exactly the same no matter what lens you use.

Wikipedia provides some good examples on Perspective Distortion and shows the math behind the Angle of View.

From that page, I like this animation of a cube that goes from extreme extension to extreme compression, when changing the focal length and distance to camera:


Image copyright 2007 SharkD, licensed CC-BY-SA 3.0

Improve this answer
\$\endgroup\$
2
  • \$\begingroup\$ Is this a "maths and logic" thing which applies to all creatures and martians that are capable of sight or is this simply a function of how the human eyes work? \$\endgroup\$
    – Pacerier
    Jan 27, 2017 at 8:00
  • \$\begingroup\$ @Pacerier It would depend on the mechanism(s) used for vision. Any vision based on an entrance pupil significantly narrower than the scene being viewed would show the effects of perspective distortion. However, if a life form had a vision system more like that of a telecentric lens viewing a scene with an FoV narrower than the size of the entrance pupil, or a vision mechanism more like a scanner than a lens, then effectively everything would be seen orthographically. \$\endgroup\$
    – Michael C
    Apr 5, 2017 at 16:25
48
\$\begingroup\$

Look up 'triple reverse zoom' or 'dolly zoom'. For a visual representation, this Wikipedia page has a decent example: http://en.wikipedia.org/wiki/Dolly_zoom

It's actually very easy to learn how this works for yourself:

  1. Find a nice bumpy object to photograph with a deep background. A person on a street is a good choice
  2. Get a zoom lens and photograph them as close as possible
  3. Back up in regular increments zooming in on each step to get a similar crop on each photo
  4. Go home and compare the photos to see the 'flattening' effect
  5. Now go back and rescue your subject...
Improve this answer
\$\endgroup\$
3
  • 1
    \$\begingroup\$ Cool! There's some great visualizations there, and that's a place I would have never thought to look. \$\endgroup\$
    – mattdm
    Dec 19, 2011 at 15:25
  • 2
    \$\begingroup\$ +1 for one good visual triumphs words, and easier to understand. \$\endgroup\$
    – Global nomad
    Jun 15, 2012 at 20:34
  • \$\begingroup\$ I've always known how this effect was achieved, but never knew the name. Thank you. \$\endgroup\$
    – byxor
    Mar 21, 2018 at 13:52
19
\$\begingroup\$

I believe the effect has to do with the RATIO of distances from the camera to various parts of the subject / scene. For example, if you take a wide-angle shot of a person's face, their features are exaggerated because the camera-to-nose distance might be half of the camera-to-ear distance.

On the other hand, consider the same shot taken with a telephoto lens (same framing / composition). In this case, you are standing further away, so the camera-to-nose : camera-to-ears distance is around 1:1. This is why telephoto shots are more flattering to the models.

Improve this answer
\$\endgroup\$
2
  • \$\begingroup\$ Does "focal length" mean the same thing as "ratio"? \$\endgroup\$
    – Pacerier
    Jan 27, 2017 at 8:02
  • 1
    \$\begingroup\$ @Pacerier - No, they are not the same. Using non-technical terms, focal length is how "zoomed in" you are. When I say "ratio," I am talking about comparing the camera-to-object distance for two objects. \$\endgroup\$
    – anon
    Feb 9, 2017 at 20:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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