Do Tilt shift lenses correct perspective or distort

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

austrokiwi1
Posts: 350
Joined: Sun Sep 14, 2014 10:53 am

Do Tilt shift lenses correct perspective or distort

Post by austrokiwi1 »

In a coin forum a leading light in coin photography has stated that shift movements cause distortion? I am a bit surprised at that view point as I have always understood that tilt and shift are used to correct perspective. I use shift in combination with long focal length lenses to allow for a pseudo axial lighting of the subject. I am honestly flummoxed when I am told that what I am doing is producing distortion

To Clarify my understanding: I have always understood that Distortion in its various types usually applies to lenses and their designs, where as Tilt and shift are means to correct issues in perspective. Can some one tell me where my understanding is wrong Thanks
Still learning,
Cameras' Sony A7rII, OLympus OMD-EM10II
Macro lenses: Printing nikkor 105mm, Sony FE 90mm F2.8 Macro G, Schneider Kreuznach Makro Iris 50mm , 2.8, Schnieder Kreuznach APO Componon HM 40mm F2.8 , Mamiya 645 120mm F4 Macro ( used with mirex tilt shift adapter), Olympus 135mm 4.5 bellows lens, Oly 80mm bellows lens, Olympus 60mm F2.8

JohnyM
Posts: 463
Joined: Tue Dec 24, 2013 7:02 am

Post by JohnyM »

Correct + with -

ChrisR
Site Admin
Posts: 8668
Joined: Sat Mar 14, 2009 3:58 am
Location: Near London, UK

Post by ChrisR »

I'm with you AK, shifts can certainly fix some distortions. But if he's a leading light, he may be thinking of different sorts of distortions. Nothing comes to mind, I must say. :?
Chris R

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

Hmm, I've never been called a "leading light" before.

Using as an example an architectural photo where the top of the building is narrower than the bottom due to converging perspective (ie, top is farther away from the camera than is the bottom), a tilt lens can be used to correct this perspective through optically "stretching" the top part of the image more than the bottom part. The image with converging perspective is "correct" as far as what the standard lens presents to the sensor. The image with "corrected perspective" is actually distorted in such a way as to make the building's walls appear vertical rather than converging.
Correct + with -


What does this mean?

JohnyM
Posts: 463
Joined: Tue Dec 24, 2013 7:02 am

Post by JohnyM »

Neglect negative value of perspective distortion by introducing positive value. Same way (different abberations type ) how eyepieces / tube lenses works in microscopes (with sole exception of Nikon).

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

JohnyM wrote:Neglect negative value of perspective distortion by introducing positive value. Same way (different abberations type ) how eyepieces / tube lenses works in microscopes (with sole exception of Nikon).
OK, makes sense.

But I don't think of an image showing converging perspective lines as being distorted. It is correct and distortion-free from the viewpoint of an observer at the reference plane of the camera/lens. Otherwise, most every image ever taken would have to be classified as "distorted". Only exception would be images of perfectly flat things taken straight on with a flat field lens. Soon as any tilt occurs, perspective comes into play. Perspective distorts the apparent shape of objects, but the resulting image of the distorted object is not itself distorted. Using a tilt lens distorts the image so that the objects appear undistorted, as if viewed straight-on with no converging perspective. It is a view that does not exist in nature, but in many cases is more aesthetically-appealing.
Last edited by ray_parkhurst on Fri Sep 09, 2016 9:06 pm, edited 1 time in total.

Charles Krebs
Posts: 5865
Joined: Tue Aug 01, 2006 8:02 pm
Location: Issaquah, WA USA
Contact:

Post by Charles Krebs »

Tilt controls the plane of sharp focus. It is shifting that plays around with the perspective.

elf
Posts: 1416
Joined: Sun Nov 18, 2007 12:10 pm

Post by elf »

I think it depends on your definition of distortion:

Code: Select all

dis·tor·tion
d??stôrSH(?)n/
noun
noun: distortion; plural noun: distortions

    1.
    the action of distorting or the state of being distorted.
    "the virus causes distortion of the leaves"
        a distorted form or part.
        "a distortion in the eye's shape or structure"
    2.
    the action of giving a misleading account or impression.
    "we're tired of the media's continuing distortion of our issues"
    3.
    change in the form of an electrical signal or sound wave 
If you think the only true view of an object is when the focal plane hasn't been shifted or tilted, then using those would be distorting. However, if you think that any plane of focus is a valid view of the object, then there is no distortion. I believe the later is correct, but if you want to settle the argument in the coin forum, you'll first have to agree on the definition.

rjlittlefield
Site Admin
Posts: 23562
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

ray_parkhurst wrote:Using a tilt lens distorts the image so that the objects appear undistorted, as if viewed straight-on with no converging perspective. It is a view that does not exist in nature, but in many cases is more aesthetically-appealing.
That depends on your definition of "in nature", I think.

If you mount a wide rectilinear lens, point the camera horizontally at the building, take a picture, and just crop appropriately, then you'll get exactly the same image that you would get with tilt/shift perspective correction from the same viewpoint.

The big advantage of tilt/shift in this case is only that it lets you fill the sensor with the building, while still keeping the camera horizontal.

--Rik

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

rjlittlefield wrote:
ray_parkhurst wrote:Using a tilt lens distorts the image so that the objects appear undistorted, as if viewed straight-on with no converging perspective. It is a view that does not exist in nature, but in many cases is more aesthetically-appealing.
That depends on your definition of "in nature", I think.
--Rik
By "in nature" I mean viewed without optical aids. Thus converging perspective is a "natural" phenomenon. You can use optical or digital techniques to stretch the image such that the converging perspective is gone, but doing this is distorting the image to fit a particular aesthetic.
Charles Krebs wrote:Tilt controls the plane of sharp focus. It is shifting that plays around with the perspective.
When I wrote "tilt" I meant tilting the object, not the lens. For instance, if you take a picture of a piece of paper that has a square drawn on it, using a lens with well-corrected geometric distortions, with the paper flat vs the sensor, the square will appear square in the image. If you then tilt the paper vs the sensor, it will take on a trapezoidal shape in the image. It may have depth of field issues, but not necessarily. A small amount of tilt, with the lens sufficiently stopped-down, may have acceptable depth of field, but will still appear distorted into a trapezoid. This is the natural, undistorted appearance of the tilted square, ie if you looked at it without camera or lens, it would be trapezoidal shaped. The image can be "corrected" such that the square appears "square" by stretching it appropriately, but this is a distortion of the image to meet a particular aesthetic.

elf
Posts: 1416
Joined: Sun Nov 18, 2007 12:10 pm

Post by elf »

ray_parkhurst wrote: When I wrote "tilt" I meant tilting the object, not the lens. For instance, if you take a picture of a piece of paper that has a square drawn on it, using a lens with well-corrected geometric distortions, with the paper flat vs the sensor, the square will appear square in the image. If you then tilt the paper vs the sensor, it will take on a trapezoidal shape in the image. It may have depth of field issues, but not necessarily. A small amount of tilt, with the lens sufficiently stopped-down, may have acceptable depth of field, but will still appear distorted into a trapezoid. This is the natural, undistorted appearance of the tilted square, ie if you looked at it without camera or lens, it would be trapezoidal shaped. The image can be "corrected" such that the square appears "square" by stretching it appropriately, but this is a distortion of the image to meet a particular aesthetic.
Actually there is no distortion involved at all. No light rays have been bent or altered. Your definition of distortion is not the same as mine. Please use another word for the phenomenon you're describing :lol:

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

elf wrote: Actually there is no distortion involved at all. No light rays have been bent or altered. Your definition of distortion is not the same as mine. Please use another word for the phenomenon you're describing :lol:
Our definitions of "distortion" are probably the same, we're just referring to two different aspects of the problem. You are correct that there is no distortion introduced during the taking of the shifted-lens image, given that the lens has sufficient coverage. But the shifted-lens image is indeed distorted compared with a non-shifted image taken from the same vantage point (Rik's wide/cropped example aside). The whole point of shifting (in architectural photography...coin photography has different considerations) is to fix (distort) the natural converging-perspective of long parallel lines so that on the image they appear parallel.

austrokiwi1
Posts: 350
Joined: Sun Sep 14, 2014 10:53 am

Post by austrokiwi1 »

Ray I really think you are applying a different definition of Distortion.

In the other forum you said this:
The methods are similar, but not the same. With pure shift, you distort the image to achieve parallax/perspective shifts. With pure tilt, you affect depth of field. A combo is good for fixing both issues. Here's a simple graphic to show the differences


You then used this picture:

Image


Now sticking to what I had be discussing which was using shift to enable lighting of a coin from a vertical position( from directly above the coin). The middle graphic of yours illustrates what I was discussing. To me that graphic shows a non distorting (ignoring any lens peripheral lens faults) set up. Everything I have done and read tells me there is no distortion involved as the coin and sensor remain parallel. In non- macro photography the same technique is used to photograph a mirror without having the photographer or camera visible in the mirror. So please explain where I am wrong and point out where the distortion is being produced as you claim is occurring. Pointing me in the direction of some on line references ( or paper references) would be helpful.

I would note in my experience that this method of lighting using shift works best with long focal length lenses at low magnification.
Still learning,
Cameras' Sony A7rII, OLympus OMD-EM10II
Macro lenses: Printing nikkor 105mm, Sony FE 90mm F2.8 Macro G, Schneider Kreuznach Makro Iris 50mm , 2.8, Schnieder Kreuznach APO Componon HM 40mm F2.8 , Mamiya 645 120mm F4 Macro ( used with mirex tilt shift adapter), Olympus 135mm 4.5 bellows lens, Oly 80mm bellows lens, Olympus 60mm F2.8

ray_parkhurst
Posts: 3416
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Post by ray_parkhurst »

This concept of shifting causing distortion versus the un-shifted image seems to be difficult for folks to understand. I have gone at this from multiple angles on both forums. For sure I am not using some arcane definition of distortion. Maybe too many words, not enough pictures...

wpl
Posts: 22
Joined: Thu Jun 28, 2012 9:43 am
Location: New Mexico, USA

Post by wpl »

Whether you call it distortion or not, the shifted image is different. In the shifted image you can see the edge of the coin. In the normal photo, the edges are invisible.

Post Reply Previous topicNext topic