Perspective (was Re: Has ultrawide angle become an overused cliche?)
in [Digital Cameras]
|
From: Neil Harrington on 12 Aug 2010 13:41 "bugbear" <bugbear(a)trim_papermule.co.uk_trim> wrote in message news:M8idnXnO9vsL7vzRnZ2dnUVZ8h6dnZ2d(a)brightview.co.uk... > Neil Harrington wrote: >> "bugbear" <bugbear(a)trim_papermule.co.uk_trim> wrote in message >> news:1KOdnRBZo9FzlfzRnZ2dnUVZ7vU3AAAA(a)brightview.co.uk... >>> Neil Harrington wrote: >>>> "Shiva Das" <shiv(a)nataraja.invalid> wrote in message >>>>> <http://www.flickr.com/photos/52865695(a)N03/4875633691/> >>>>> >>>>> All I did was crop the 11mm photo to the same field of view as the >>>>> 35mm >>>>> photo. There is absolutely no difference in the perspective. >>>> Cropping the photo changes its perspective. >>> OK. Would you please define, as clearly, carefully and explicitly >>> as you can, what you mean by "perspective" >> >> <groan> >> >> If we're going to start getting into arguments about lengthy definitions >> of rather ordinary terms, this will have to wait until I get back as I'm >> going out of state in a little while. >> >> But briefly, in the sense we are using the term here, perspective is the >> rendering on a two-dimensional surface of the shapes and apparent >> relationship of objects (or parts of a single object) in a >> three-dimensional world. > > Excellent. > > Now how in $DEITY's name would cropping alter that? I have already answered this in more than one previous post. I wish you would read all the posts so I don't have to repeat the same thing over and over. But I will again, this one time. (If the same question keeps coming up I may make a boilerplate for it.) Perspective is a quality of the ENTIRE PICTURE, not one small part. All "proofs" that perspective is solely a function of viewpoint (and that focal length has nothing to do with it) are based in one way or another on resizing the central portion of the picture only, as if the center and only the center established perspective for the whole. Obviously, magnifying or reducing one part of the picture does not (and cannot possibly) change the perspective. Here's a reductio ad absurdum question for you. If only a small part of the center were important to perspective, why not make that part smaller still? Why not just keep shrinking it? Suppose at the very center of the picture there is a pea -- if we should go so far as to crop away everything except that one little pea, would that picture of the pea still convey the full perspective of the original picture? Wide-angle pictures often contain important elements of perspective that simply are not there in long-lens pictures of the same subject taken from the same viewpoint. And those elements of perspective near the edges of the wide-angle picture CANNOT BE DUPLICATED with a long lens. This is of course assuming there are three-dimensional objects there that can contribute to perspective. (A wide-angle picture of blank sky would contain no perspective at all.) It is because of these differences in perspective that are unique to wide-angle lenses that we can immediately tell a wide-angle picture when we see one. If those important differences in perspective were not there, we would not be able to tell a wide-angle picture from a normal or long-lens one.
From: Neil Harrington on 12 Aug 2010 14:03 "Savageduck" <savageduck1@{REMOVESPAM}me.com> wrote in message news:201008100856127987-savageduck1(a)REMOVESPAMmecom... > On 2010-08-10 08:27:54 -0700, "David Ruether" <d_ruether(a)thotmail.com> > said: > >> >> "Peter" <peternew(a)nospamoptonline.net> wrote in message >> news:4c615c9c$0$5496$8f2e0ebb(a)news.shared-secrets.com... >>> "David Ruether" <d_ruether(a)thotmail.com> wrote in message >>> news:i3ppsj$3ln$1(a)ruby.cit.cornell.edu... >> >>>> Ah, but the brain *doesn't* correct it to rectilinear perspective! >>>> THAT'S what's so much fun to show people (that the way they >>>> think they see isn't the way they really do see...;-). >> >>> IIRC There have been some experiments where people were fitted with >>> prism glasses to make everything appear upside down. The >>> glasses were left on for some period of time. After a while the upside >>> down illusion corrected itself and when the glasses were >>> removed, everything appeared upside down again. >>> I am going from memory and too lazy to find the paper. >>> >>> <good stuff snipped> >> >> <good DR stuff restored...;->: >> >> "Ah, but the brain *doesn't* correct it to rectilinear perspective! >> THAT'S what's so much fun to show people (that the way they >> think they see isn't the way they really do see...;-). There are >> simple exercises that permit people to attend to their off-center >> field of vision, and once that is accomplished, it is easy to see under >> some conditions of high color or tone contrast the clear curvatures >> of the fisheye perspective type. Also, logically there can be no >> 180-degree rectangular perspective image (the image plane would >> be infinitely large, and the "lens" FL would be zero), although we >> do see slightly more than 180 degrees horizontally, so there cannot >> be anything near a 180-degree fisheye-to-rectangular perspective >> conversion. For more, with an image approximation of how we do >> see, go here -- >> http://www.donferrario.com/ruether/seeing_and_perspective.htm >> (There are more articles on perspective, here -- >> http://www.donferrario.com/ruether/articles.html .)" >> >>> -- >>> Peter >> >> Yes - the brain can learn to apply *possible* corrections >> (exchanges of up and down, left and right orientations - and >> to synthesize missing parts in the visual field, which we all have), >> but not impossible things, such as converting a 180+ degree >> fisheye view into a 180+ degree rectangular-perspective view. >> Partly (maybe mostly...;-) why most people assume that they >> see in rectangular perspective is that most people visually >> "attend" to only a narrow angle of view at any one moment >> and "construct" their impression of visual reality from a large >> series of such narrow views - and within a narrow angle of >> view, all the several possible perspective types look nearly >> the same... >> --DR > > ...and then there is telephoto compression when viewpoint/camera/subject > "perspective" is maintained. I think a big point of contention is > understanding that a change in how an image "looks" because of a change in > FL, or sensor size is not a change of "perspective." > This is an error I have been guilty of. > > Strictly speaking perspective is related to viewpoint/camera/subject > position. Where lens induced distortion, and compression, with maintained > viewpoint/camera/subject position, is just that, a change to the > image/scene from the same perspective. > So one could say you can get several different views of a single scene > from the same perspective, with lenses of different FL, or even with > different cameras, provided the subject to sensor plane is maintained as a > constant. That is a slightly different use of the term "perspective," though. When for example you speak of "from my perspective," you mean "from my point of view," etc. That's a correct but different usage from "perspective" as in perspective drawing. In such drawing the perspective is incorporated in the drawing itself. You can change your viewpoint relative to the drawing and its perspective does not change. (Think of a portrait in which the eyes are always looking right at you, no matter where you stand.) The spatial compression and "distortion" that you mention in the case of long-lens shots is, in fact, not distortion at all even though it has the appearance of distortion. It looks wrong because it approaches isometric drawing, which is correct for isometric drawing but "wrong" in terms of the way we see things in the real world, which is to say, in perspective. The longer the lens relative to the format, the closer it approaches isometric, though of course it can never actually reach isometric. But the *perspective* of a long lens is correct even if it looks wrong, and is what it has to be. Likewise, the perspective of a wide-angle lens is (assuming a good rectilinear lens of course) what it has to be. It isn't really wrong or distorted even at the edges where it looks "wrong" because objects are stretched radially.
From: Paul Furman on 12 Aug 2010 14:05 Neil Harrington wrote: > "bugbear"<bugbear(a)trim_papermule.co.uk_trim> wrote in message > news:M8idnXnO9vsL7vzRnZ2dnUVZ8h6dnZ2d(a)brightview.co.uk... >> Neil Harrington wrote: >>> "bugbear"<bugbear(a)trim_papermule.co.uk_trim> wrote in message >>> news:1KOdnRBZo9FzlfzRnZ2dnUVZ7vU3AAAA(a)brightview.co.uk... >>>> Neil Harrington wrote: >>>>> "Shiva Das"<shiv(a)nataraja.invalid> wrote in message >>>>>> <http://www.flickr.com/photos/52865695(a)N03/4875633691/> >>>>>> >>>>>> All I did was crop the 11mm photo to the same field of view as the >>>>>> 35mm >>>>>> photo. There is absolutely no difference in the perspective. >>>>> Cropping the photo changes its perspective. >>>> OK. Would you please define, as clearly, carefully and explicitly >>>> as you can, what you mean by "perspective" >>> >>> <groan> >>> >>> If we're going to start getting into arguments about lengthy definitions >>> of rather ordinary terms, this will have to wait until I get back as I'm >>> going out of state in a little while. >>> >>> But briefly, in the sense we are using the term here, perspective is the >>> rendering on a two-dimensional surface of the shapes and apparent >>> relationship of objects (or parts of a single object) in a >>> three-dimensional world. >> >> Excellent. >> >> Now how in $DEITY's name would cropping alter that? > > I have already answered this in more than one previous post. I wish you > would read all the posts so I don't have to repeat the same thing over and > over. But I will again, this one time. (If the same question keeps coming up > I may make a boilerplate for it.) > > Perspective is a quality of the ENTIRE PICTURE, not one small part. All > "proofs" that perspective is solely a function of viewpoint (and that focal > length has nothing to do with it) are based in one way or another on > resizing the central portion of the picture only, as if the center and only > the center established perspective for the whole. Obviously, magnifying or > reducing one part of the picture does not (and cannot possibly) change the > perspective. > > Here's a reductio ad absurdum question for you. If only a small part of the > center were important to perspective, why not make that part smaller still? > Why not just keep shrinking it? Suppose at the very center of the picture > there is a pea -- if we should go so far as to crop away everything except > that one little pea, would that picture of the pea still convey the full > perspective of the original picture? > > Wide-angle pictures often contain important elements of perspective that > simply are not there in long-lens pictures of the same subject taken from > the same viewpoint. And those elements of perspective near the edges of the > wide-angle picture CANNOT BE DUPLICATED with a long lens. This is of course > assuming there are three-dimensional objects there that can contribute to > perspective. (A wide-angle picture of blank sky would contain no perspective > at all.) > > It is because of these differences in perspective that are unique to > wide-angle lenses that we can immediately tell a wide-angle picture when we > see one. If those important differences in perspective were not there, we > would not be able to tell a wide-angle picture from a normal or long-lens > one. This seems like mostly an argument about semantics but here's an example that supports the idea of real differences in wide angle 'perspective'. Take a pic with a super-wide, then crop the *corner* of the image. Now try to reproduce that perspective with a telephoto lens. It could be accomplished with a stitched pano, if the stitching software stretches things but not without software distortion. Check the drawing on the bottom of this page, of a distorted sphere in the cropped corner of a wide angle view: http://www.donferrario.com/ruether/lens-angle-of-view-and-perspective.htm I don't think that could be reproduced with a telephoto lens. Put a striped ball on the left side of your desk and try it.
From: Neil Harrington on 12 Aug 2010 14:12 "tony cooper" <tony_cooper213(a)earthlink.net> wrote in message news:q4p36696nv8l3oisef0i1ld2feo29krg7f(a)4ax.com... > On Tue, 10 Aug 2010 09:57:23 -0400, "Peter" > <peternew(a)nospamoptonline.net> wrote: > >>"Savageduck" <savageduck1@{REMOVESPAM}me.com> wrote in message >>news:2010080902311316807-savageduck1(a)REMOVESPAMmecom... >> >>> >>> A valuable lesson learned with my own material. >>> >>> Hoisted on my own petard so to speak. ;-) >>> >> >>According to my understanding it means you smelled your own farts. >> >>http://www.thefreedictionary.com/petard > > Note that the actual phrase is "hoist with one's own petard". "With", > not "on". The exploding bomb hoists one. > > Shakespeare used the phrase in "Hamlet", but it's spelled "petar" in > some versions of the work. Yes, I believe that's the way he wrote it. (Some helpful souls may have corrected his spelling in versions long after his death.) I assume he wrote it "petar" because he was going by the French pronunciation, which may also have been the way it was pronounced in English at that time. The last consonant in a word in French is usually not pronounced unless it's c, r, f or l. (That may be the only rule I still remember from my old French classes.)
From: Neil Harrington on 12 Aug 2010 14:54
"Paul Furman" <paul-@-edgehill.net> wrote in message news:AKCdnTDHmc6opfnRnZ2dnUVZ5oOdnZ2d(a)giganews.com... > Neil Harrington wrote: >> "bugbear"<bugbear(a)trim_papermule.co.uk_trim> wrote in message >> news:M8idnXnO9vsL7vzRnZ2dnUVZ8h6dnZ2d(a)brightview.co.uk... >>> Neil Harrington wrote: >>>> "bugbear"<bugbear(a)trim_papermule.co.uk_trim> wrote in message >>>> news:1KOdnRBZo9FzlfzRnZ2dnUVZ7vU3AAAA(a)brightview.co.uk... >>>>> Neil Harrington wrote: >>>>>> "Shiva Das"<shiv(a)nataraja.invalid> wrote in message >>>>>>> <http://www.flickr.com/photos/52865695(a)N03/4875633691/> >>>>>>> >>>>>>> All I did was crop the 11mm photo to the same field of view as the >>>>>>> 35mm >>>>>>> photo. There is absolutely no difference in the perspective. >>>>>> Cropping the photo changes its perspective. >>>>> OK. Would you please define, as clearly, carefully and explicitly >>>>> as you can, what you mean by "perspective" >>>> >>>> <groan> >>>> >>>> If we're going to start getting into arguments about lengthy >>>> definitions >>>> of rather ordinary terms, this will have to wait until I get back as >>>> I'm >>>> going out of state in a little while. >>>> >>>> But briefly, in the sense we are using the term here, perspective is >>>> the >>>> rendering on a two-dimensional surface of the shapes and apparent >>>> relationship of objects (or parts of a single object) in a >>>> three-dimensional world. >>> >>> Excellent. >>> >>> Now how in $DEITY's name would cropping alter that? >> >> I have already answered this in more than one previous post. I wish you >> would read all the posts so I don't have to repeat the same thing over >> and >> over. But I will again, this one time. (If the same question keeps coming >> up >> I may make a boilerplate for it.) >> >> Perspective is a quality of the ENTIRE PICTURE, not one small part. All >> "proofs" that perspective is solely a function of viewpoint (and that >> focal >> length has nothing to do with it) are based in one way or another on >> resizing the central portion of the picture only, as if the center and >> only >> the center established perspective for the whole. Obviously, magnifying >> or >> reducing one part of the picture does not (and cannot possibly) change >> the >> perspective. >> >> Here's a reductio ad absurdum question for you. If only a small part of >> the >> center were important to perspective, why not make that part smaller >> still? >> Why not just keep shrinking it? Suppose at the very center of the picture >> there is a pea -- if we should go so far as to crop away everything >> except >> that one little pea, would that picture of the pea still convey the full >> perspective of the original picture? >> >> Wide-angle pictures often contain important elements of perspective that >> simply are not there in long-lens pictures of the same subject taken from >> the same viewpoint. And those elements of perspective near the edges of >> the >> wide-angle picture CANNOT BE DUPLICATED with a long lens. This is of >> course >> assuming there are three-dimensional objects there that can contribute to >> perspective. (A wide-angle picture of blank sky would contain no >> perspective >> at all.) >> >> It is because of these differences in perspective that are unique to >> wide-angle lenses that we can immediately tell a wide-angle picture when >> we >> see one. If those important differences in perspective were not there, we >> would not be able to tell a wide-angle picture from a normal or long-lens >> one. > > This seems like mostly an argument about semantics but here's an example It is "about semantics" insofar as the word "perspective" has different shades of meaning. But for people to deny that wide-angle perspective exists when they can clearly *see* wide-angle perspective and recognize it when they see it, is rather astounding. > that supports the idea of real differences in wide angle 'perspective'. > Take a pic with a super-wide, then crop the *corner* of the image. Now try > to reproduce that perspective with a telephoto lens. It could be > accomplished with a stitched pano, if the stitching software stretches > things but not without software distortion. > > Check the drawing on the bottom of this page, of a distorted sphere in the > cropped corner of a wide angle view: > http://www.donferrario.com/ruether/lens-angle-of-view-and-perspective.htm > I don't think that could be reproduced with a telephoto lens. > Put a striped ball on the left side of your desk and try it. Exactly right. You could not reproduce the corner of a wide-angle shot with a long lens, assuming the w.a. shot contained some three-dimensional object(s). |