About perspective distortions. Note from reader Radozhiva

Material on Adobe Lightroom Classic specially for Radozhiva prepared Alexander Onishchenko (facebook).

Fig. 1

Fig. 1

Good time, readers of Radozhiva! I am with you, Alexander Onishchenko.

The promised post detailing the interface and operation of the Luminance Range and Color Range tools in Adobe Lightroom 11 will follow next week. Today, let's talk about perspective distortion when shooting portraits.

Most self-taught photographers, and the graduates of countless photographic schools who have joined them, swear by all the megapixels of their cameras that the shorter the focal length of the lens, the more it distorts the perspective. They often illustrate their statements with a set of portraits like the one shown here. Why do they think so? - Yes, because they were taught that way. Or they Google it themselves.

Indeed, there are many examples on the net, like an animated slide from a series of pictures from Dan Vojtech (Fig. 2), which, it would seem, clearly confirm this relationship.

Fig. 2

Fig. 2

But in reality, not everything is as simple as it really is. We will argue on the static version of the portraits (Fig. 3). It contains portraits taken at various focal lengths (from 20 to 200 mm). If you carefully compare the individual frames, you can see that they differ noticeably. In short-focus shots, we observe the so-called. the “horse face” effect, when the model's face is unreasonably stretched forward. As the focal length lengthens, this effect weakens, and the face, on the contrary, “unfolds” along the frontal plane:

Fig. 3

Fig. 3

Why am I questioning what our eyes confirm? Because, strictly speaking, perspective distortion does not depend in any way on the focal length. The fact is that all 9 versions of the portraits were shot “in one size”. Those. the vertical size of the portrait always occupies the same part of the frame. And this is fundamentally important!

The longer the focal length of the lens, the further the photographer has to move away from the model in order to maintain this specified size. And it is the change in the distance to the model that affects the perspective distortion.

It is not difficult to verify the validity of this statement: it is enough to take a similar series of pictures (at different focal lengths) on the spot. Yes, in the resulting frames, the face size will be significantly different (the shorter the focal length, the smaller the face). But if we then bring all the frames to a single scale in Photoshop, we will see that the perspective distortions on them are exactly the same:

Fig. 4

Fig. 4

So, perspective distortion does not depend on the focal length of the lens, but only on the distance between the camera and the model. When shooting in one size, changing the focal length forces the photographer to change the shooting distance, and this is what changes the apparent perspective.

You will find more materials from readers of Radozhiva here.

Add a comment: Oleg Shevchenko

 

 

Comments: 49, on the topic: About perspective distortions. Note from reader Radozhiva

  • Michael

    Well, this is, in principle, understandable. Otherwise, 35mm on the crop would give different distortions than 50mm on the FF. Another thing is that super-widths still like to pull corners. And wide shooting will somehow capture the foreground and show the very distortions. And so, for a photo of groups of people 85mm is not necessary, of course.

    • ROMAN

      It is necessary / not necessary - it's an amateur's business.
      But unambiguously 85 group photos will look much better than wide ones) Another thing is that it is not always possible to shoot a long focus.
      But personally, I always shoot group photos exactly at 85, also on the open one, when there is such an opportunity. Photos with the resulting compression of space come out very beautiful. And when you give it to the customer, then he will definitely not have a thought, they say, "I would have taken it on the phone."

      • Eugene o

        Shake your hand. I also like to shoot group photos at 85k, people have normal-shaped faces at the edges of the frame, but one thing - you have to move away VERY far. And 10 meters, and 15 is not the limit.

      • QWERTY

        I know one photograph, because she basically does not need aperture, she shoots everything in the open. This is a lot for everyone, but people like all kinds of rubbish 😂🤣 Compression of perspective kills the semantic component of the picture - just a group of people, and where and when it was filmed is unclear, whether at home in the courtyard, or in the Kremlin. And the soap around the edges is still something. Bokeh is a good thing, but not in the brain. As for the author of the article, he went to the wrong steppe. The shorter the focal length, the closer you need to get to the subject. Accordingly, if you shoot wide, then the difference in distances from the plane of the matrix to the nose and ears will be quite large compared to the distance to the subject. Therefore, we get a snobel and small ears, that is, the notorious "volume". With an increase in FR, the shooting distance also increases, and, accordingly, the difference in the distance between the nose and ears decreases in comparison with the distance to the object, so the image is “flatter”

        • Seladir

          As a result, you described the influence of perspective as the author, only without using this word, in a simple way, but the essence is the same)

          • QWERTY

            what the respected afftor wrote about has a purely academic interest. And corny it is obvious

    • Vitaly Pinchuk

      Therefore, the article is called "perspective distortion", and not distortion from optics.

      • Alexander Onishchenko

        I wanted to explicitly write this idea in the publication, but decided that it was already clear.

        • Andrei

          It looks more like it was done on purpose - for the sake of srach.

      • gene

        Because an article about perspective distortions, about how one walks with one’s feet to get a picture, and not about the distortion of optics (about how optics creates field curvature), would teach materiel-shame. I heard a ringing and I don’t know where it is, I’m selling 150 photographs for 15 thousand rubles, art! Why did they show you Figure 4? In order to explain that it’s not a matter of distortion.

        • Rodion

          Well, you are confusing field curvature and distortion. It’s up to you to teach the materiel too, which means you need to be sent.

  • Oleg

    But at a wide angle, the plump ones become thin

    • Lord toilet

      And the foreheads are pulled over the back of the head)

      • Michael

        But the chest is larger)

  • Alexey

    This law is described in the book by L.P. Dyko "Photocomposition", chapter 4 "Visual tasks in photography" (edition 1962, page 80):
    "The perspective of a photographic image in shots taken with different lenses, but from the same point, remains the same, does not change."

    • Alexander Onishchenko

      And not only her. But yes - Lydia Pavlovna wrote a very useful book.

  • ROMAN

    By the way, I did the portrait of a bear. One photo at 85mm, another photo at 28mm.
    Perspective, compression, notorious proportions have not changed at all.
    However, even after illustrative examples, many amateur photographers and “professionals” cannot get the most basic things into their heads. Something to explain has long been tired.

    • Nicholas

      The difference is in the resolution of these two images. If at 85 mm this is a real shot (for example), then at 28 it is already a wild crop. And it's good if you have at the same time a 61MP (at least 45MP) matrix with a 28mm lens corresponding to this matrix (in terms of resolution)…. And in another way - soap. I think this is the best explanation about "the most basic things"
      Otherwise, everything is correct. Comrade Trueash gave a very good link on this issue.

      • Andrei

        I took it off my tongue. Indeed, let's do 500mp each in cameras and then we will shoot everything wide and scattered, the most nonsense article that I have seen lately. It is logical to think that if you move a couple of steps away from the model in width, then perspective distortions are leveled by the distance. What a surprise! And after that, blame some schools and the like ... Dislike, if only it were possible here.

        • Oleg Shevchenko

          Alexander Onishchenko would like to give advice to stay in his topic - Lightroom, in which he (I hope) understands more.

        • ROMAN

          Everything is logical, everyone knows everything, but then the traditional crap developed in the comments)
          This means that not everyone knows and understands everything)

          • Andrei

            There are simply people who do not distinguish perspective from optical (distortion). This is completely different. What was the subject? In promising. Done

        • Seladir

          If you saw the call to shoot like this in a purely illustrative example, then these are problems of your perception, not articles. People still stupidly regularly on this topic, the comments here confirm this.

          • Victor

            +1

            Until now, there are those who claim that "here is de 85 a real panteret, and 56mm (even when installed on a crop) is NED, for DISTORTIONS !! 11"

          • Andrei

            You saw this call in my message, but for some reason say that I saw it, although on my part it was just irony. Difficult, try to re-read my comment a couple more times ... And also, to be stupid in understanding perspective distortions or, all the same, to confuse them with optical ones? Here is an article in general about something else, I advise you also not to be blunt and not to draw pussy to your nose (one topic to another), and also, like those who “blunt”, be guided by the author's mistake, when writing this subject - do not specify what it is refers specifically to perspective distortions, which should not be confused with optical. 🤦‍♂️

  • Trueash

    FStoppers has a good video on the same topic:
    https://youtu.be/_TTXY1Se0eg

  • Eugene o

    Well, technically there should still be a slight difference, just because of the tiny size of the front glass in relation to the objects being filmed, it can be neglected. If the glass were human-sized in diameter, perspective distortions would appear. At least it seems logical from the point of view. geometry.

  • Michael

    Since I am an applied geometer, giving, among other things, a course of lectures on geometric perspective for students-architects, I can confidently say that geometric perspective distortion is significantly more dependent on the diameter of the optical elements of the lens. It is enough to take two lenses with the same focal length, but different lens diameters - for example, Nikkor 18-55 and the same Nikkor 18-105. If the first one "at the short end" builds vertical lines along the edges of the frame almost straightforward, then the second one quite sensitively "rounds" the frame ...

    • Dmitriy

      Article for perspective distortion. The fillets you are writing about are not perspective distortions, are they? This is distortion, i.e. optical distortion.
      And one more thing: I had Nikkor AF-P 18-55 and Nikkor 18-105 right at hand.
      The screenshot shows on the top frame - 18-55, on the bottom - 18-105.
      Photographed from the same distance and equally perpendicular to the wall.
      All distortion corrections are disabled. On both, you can see the rounding of the lines, but as for me, so on the bottom (18-105) - a little straighter than on the top (18-55).

  • Oleg Shevchenko

    "Perspective distortion depends ... only on the distance camera - model" - no, not only. It also depends a lot on the optical design of the lens.
    What kind of dilettante nonsense are you publishing here?

    • Victor

      Then tell us how perspective (!) Distortion depends on the optical design of the lens. It would be very interesting to listen to :-)

      • Oleg Shevchenko

        Google and read about CRC (Close Range Correction)

        • Victor

          Even more interesting.

          Then tell us how the Nikon CRC system relates to the level of perspective distortion)))

          Or have you confused it with the Nikon PC?

    • Alexander Onishchenko

      1. It seems, Oleg, you are the only one who confuses optical distortions introduced by the lens design with perspective ones.
      2. The proposed level of discussion does not allow me to continue communicating with you.

      • Oleg Shevchenko

        Not optical, you are our lightrumman, but perspective distortions. Well, okay, a fool will always stand his ground. I am also not interested in conducting discussions with you.

    • Dmitriy

      Perhaps Oleg Shevchenko wants to make a revolution in the concept of "Perspective distortion" and their relationship with lenses. At least, I have never come across any substantiated statements that perspective distortion can depend on the optical design of the lens. Well, perhaps from the words of some few people who cannot confirm their words.
      As for the Nikon CRC system, there is also not a word that this system in any way affects the transmission of perspective distortions. There we are talking only about improving the quality of the image formed by the lens, reducing the impact on the image quality (and not on the transmission of perspective distortions) of the focusing distance.

    • Dmitriy

      Oleg Shevchenko, plunge first into the definition of “Perspective distortions”.
      For example here: https://360wiki.ru/wiki/Perspective_distortion_(photography)
      The screen shows the nuance of the independence of perspective distortions from the lens that interests you.
      And those distortions that the lens can give are already optical distortions.
      On my own I would add that perspective distortions are caused exclusively by the geometric laws of mapping 3-dimensional space onto a plane (camera matrix). Anything that a lens can add to these laws is called optical distortion.

  • Oleg Shevchenko

    It's just that earlier it was interesting and informative on joyful life, and recently, thanks to the work of such onishchenko, it was sadness.

    • Dmitriy

      On the contrary, I like that she is joyful, unlike other thematic resources, not like all sorts of “smoking rooms” and the chatter of “specialists” on benches with sunflower seeds. True, sometimes there are exceptions, you have proven it.

    • Arkady Shapoval

      Reviewed here today 300 / 2.8 G II VR, with examples on Nikon D40 - in the best traditions of Radozhiva :)

  • Alexey

    Good article, it was interesting.

  • Onotole

    As entertainment, I will refute the refutation of the thesis that “the shorter the focal length of the lens, the more it distorts the perspective”))
    Since the essence of the refutation in the article boils down to the fact that the plaster head shot at different focal lengths does not change the perspective in any way if the frames are cut to the same size. HOWEVER, the author deliberately or accidentally forgets that if something is sprinkled, then the EFR will inevitably change. Moreover, it is necessary to sprinkle exactly as many times as the focal length has been shortened.
    So it turns out that the thesis "the shorter the focal length of the lens, the more it distorts the perspective" - ​​is correct, only for absolute fidelity it needs to be slightly corrected "... the shorter the equivalent focal length ..."

    • Dmitriy

      "As entertainment - I will refute once again ..."
      No, they did not refute it. Lenses with any EGF (i.e., lens + matrix size) will give the same perspective distortion provided the same distance to the subject.
      If you provide the same framing with lenses with different EGF, then you will inevitably have to change the distance to the subject, which, of course, will lead to a change in the influence of perspective distortions. But within the framework of this article, we are not interested in the condition of the same framing.
      It is argued that for any situation where the distance to the subject is the same, perspective distortion will be the same regardless of the lenses (and matrices as you wanted to argue).

      • Andrei

        Well, yes, well, yes 👍 But for some reason, the author at the end adds a gif with a gi.head "for a visual example", which supposedly does not distort. Although he actually cropped all the photos, except for the telephoto lens, this leads us to the EGF, because this is how the CROP matrices work - “But if then in Photoshop we bring all the frames to a single scale ...”. The author of this article misled no less than people claiming that it is “impossible” to take pictures of portraits.

    • Alexander Onishchenko

      Equivalent focal length indicates the focal length a lens would have for the widespread 24x36mm frame size (35mm film frame or “full frame” digital sensor with crop factor 1) that would produce an image with the same viewing angles.

      When the frame has already been taken (as in our case), and is simply scaled in Photoshop while maintaining the proportions, there are no longer any focal points for it. It is just an image that can be captured by the camera or drawn by hand.

      “… The author intentionally or accidentally forgets” - smiled from ear to ear: o)

      • Andrei

        Bulletproof! It is strange that you ignore the logic in his words, i.e. if you are already cropping a captured image to a width of width before getting a head of the required size, then you just get the EGF of the telephoto lens for a specific matrix size (much smaller size). You seem to understand, but for some reason you are misleading the people around you (read as you are deceiving) with your “experiments” in elementary things (and I don’t take those people who confuse perspective with optical). The picture at the beginning of this article by Dan Vojtech shows the whole point and is correct. Done

  • ROMAN

    Sergei Evgenievich Medynsky - Soviet documentary cinematographer, director, professor of the department of cinematography at VGIK. Honored Artist of the Russian Federation. Lenin Prize Laureate.

  • Kirill

    I approve of such articles. All the same, this is some kind of literacy that allows you to use perspective distortions as an artistic tool.

    Surprised of course that someone does not agree, even in spite of the examples given.

    • Nicholas

      Alexander, having re-read the article again (after reading the comments), would like to wish you to continue to do this business - to eliminate illiteracy in photography and in photosoft :). And do not pay attention to the "super-intelligent and knowledgeable" theorists. As a rule, aggressiveness is a sign of narrow-mindedness, but with a desire to assert itself. Continue your business. Good luck

Add a comment

Copyright © Radojuva.com. Blog author - Photographer in Kiev Arkady Shapoval. 2009-2023

English-version of this article https://radojuva.com/en/2021/11/correct/?replytocom=507803

Version en español de este artículo https://radojuva.com/es/2021/11/correct/?replytocom=507803