Commons:Featured picture candidates/File:West Pier April 2018 04.jpg

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search

File:West Pier April 2018 04.jpg, not featured[edit]

Voting period is over. Please don't add any new votes.Voting period ends on 1 Jan 2019 at 11:23:47 (UTC)
Visit the nomination page to add or modify image notes.

New and old monuments on the Brighton seafront
  • Wrt @Spurzem, Ikan Kekek, and Daniel Case: this is what Daniel refers to "real life". This isn't camera distortion; it is what you see when you look up. You don't pay much attention to it and the edges of your vision are kinda blurry, but the verticals converge in reality too (just as parallels on the ground converge when you look down a road or railway track into the distance). The difference is that this entire scene is captured and frozen with the camera looking up at the capsule, whereas if you were standing there, as you looked at the distant buildings, your gaze would shift to a more level viewing direction, and they'd no longer slope. And as you gazed up and down the pillars, in 3D, shifting focus all the time, your brain would be constantly saying to you "It's ok, these are actually straight up and down". And since your retina is curved, your brain is constantly straightening out those bent lines your eye produces. A more representative equivalent to reality would be one of those 360° viewers that let you shift your gaze around a sphere, and where the edges of the frame are wonky but but straighten out as you shift them to the centre of the viewport. The perspective here is correct. It is the perspective one gets standing on the beach looking up at the capsule.
We may be trained that buildings should be rendered in 2D with parallel vertical lines, but that is only an approximation to reality, and only looks realistic if the vertical angle of view is modest -- which is best achieved from a distance and from a height. The only way you will get a correct perspective of this scene, with verticals going straight up your monitor screen, and the capsule faithfully rendered how the eye sees it, is if Arild bought a drone and took the photo at height and some way out to sea. The vertical angle-of-view here is simply too great for the standard rectilinear perspective to look "normal" (remember that nearly half the vertical angle of view, when standing on the beach, is "wasted" going into the sand). The effect of "correcting" the verticals does weird things to the base of the capsule and it starts looking like it is tilted 45°. We sometimes see this when Diliff includes too much ceiling in one of his cathedral interiors, and round discs on the ceiling end up looking like they are angled and heavily distorted. The "corrected" version looks like computer generated art by someone who hasn't yet grasped how perspective and viewpoint works. Please just oppose because you don't like it, that you wish it had been photographed from further back and further up, and stop trying to claim the laws of physics need fixing. -- Colin (talk) 08:51, 27 December 2018 (UTC)[reply]
I stated my case above, and I didn't mention physics once. -- Ikan Kekek (talk) 08:54, 27 December 2018 (UTC)[reply]
Charles, binocular vision makes no different to the vertical field of view, or to how parallels converge with distance. If your point is that our eyes move about as we gaze on different parts of the scene, then yes, that is a difference from a wide angle still picture on your monitor. It doesn't change the fact that verticals converge when you look up from the ground near a tall building. -- Colin (talk) 11:04, 27 December 2018 (UTC)[reply]
Ikan, you agreed with Lothar, that it "needs perspective correction", which implies the perspective is wrong, faulty, and could be fixed. Let's imagine a standard architectural perspective taken from this viewpoint on the sand. According to Google Maps, Aldrid is 50m from the base of the i360, which is 162m tall. If we pretend Aldrid is 2m tall and holding the camera to his eye, then we have a right-angled triangle 50m base and 160m tall with an angle of 73° at the camera. To get the vertical angle-of-view, we double that. So 146°. The widest ever rectilinear lens is 126° diagonal angle-of-view, much less vertically. Anything above 100° is heavily distorted. It is not "correct". Compared to 73°, the maximum the human eye can see above the horizontal, if looking straight ahead, is about 50° but that isn't even clear or in colour. We can only see in colour up about 25° and much less than that clearly. Any human standing here, hoping to see the top of the i360, is looking up with their head tilted back. Demanding a standard architectural perspective with a vertical field of view of 146° is just crazy man. The only fix that can be done here, to get the kind of image some prefer/want, is to get much much further back and ideally about 80m up in the sky. -- Colin (talk) 11:04, 27 December 2018 (UTC)[reply]
As an example of a standard architectural photo, File:Mount Stuart House 2018-08-25.jpg is I believe about 25m tall and I was 160m distant when photographing it. So the vertical angle of view of that photo is more like 20°. -- Colin (talk) 11:30, 27 December 2018 (UTC)[reply]
Here's how the BBC reported the i360 -- a "Getty Images" picture taken from even closer with even more sloping verticals. Brighton's own paper here -- with a "Press Association" image taken from even closer to the ground. And in today's Guardian newspaper, this photo of the Houses of Parliament with gloriously sloping verticals. Taken by Andy Rain, chief regional photographer for the European Press Agency. Guess we should ping him an email about his newbie mistake. -- Colin (talk) 13:31, 27 December 2018 (UTC)[reply]
What I said is that I prefer the "corrected" version. I simply like it better, although I'm not sure I'd vote to feature it, if nominated. And I find this degree of slant to the buildings strange to look at. It's an aesthetic reaction. It's totally fine to disagree with it, as you do, and obviously, there are other professional photographers who do, but that's what it is. No weighty principles are involved on my end. -- Ikan Kekek (talk) 02:40, 28 December 2018 (UTC)[reply]
Ikan thanks for the explanation, and I agree with your usage here: that when the photographer is closer to the subject than the subject is tall, then "correction" should be used within quote marks. -- Colin (talk) 10:02, 28 December 2018 (UTC)[reply]
Fair enough, but maybe that means that in my opinion, an FP from this angle might not be possible. And if so, it wouldn't be the first time a particular view of a particular motif were deemed not to be featurable. -- Ikan Kekek (talk) 19:10, 28 December 2018 (UTC)[reply]
Well much of composition is finding the correct position from which to take a photograph, so not surprising that many positions aren't regarded as ideal or best. My argument has simply been that if you don't like the results here, the solution is only to re-take it from much further back (and perhaps higher up). It can't be made "correct" in software; the laws of physics and mathematics don't allow it. -- Colin (talk) 19:20, 28 December 2018 (UTC)[reply]
Understood. -- Ikan Kekek (talk) 21:46, 28 December 2018 (UTC)[reply]
Confirmed results:
Result: 7 support, 5 oppose, 0 neutral → not featured. /MZaplotnik(talk) 15:02, 1 January 2019 (UTC)[reply]
Retrieved from "https://commons.wikimedia.org/w/index.php?title=Commons:Featured_picture_candidates/File:West_Pier_April_2018_04.jpg&oldid=333052689"