Commons:Featured picture candidates/File:St Stephen Walbrook Church Interior 1, London, UK - Diliff.jpg

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

File:St Stephen Walbrook Church Interior 1, London, UK - Diliff.jpg, featured[edit]

Voting period is over. Please don't add any new votes.Voting period ends on 26 Jan 2016 at 09:31:25 (UTC)
Visit the nomination page to add or modify image notes.

St Stephen Walbrook
  • In my opinion, the distortion is too strong for FP. I can see the distortion just by looking at the thumbnail. I think the distortion might be fixable with software. --Pine✉ 18:53, 20 January 2016 (UTC)[reply]
  • Did you actually read what I said though? I don't believe what you are describing is distortion. I think it's the actual physical properties of the pillars, where they taper at the top. This is normal in classical architecture. Diliff (talk) 22:44, 20 January 2016 (UTC)[reply]
  •  Comment I doubt it. I see what User:Pine sees, too. The straight verticals of the pillars are not completely straight, but they do look to my eye, anyway, to be of consistent size from the top to bottom of their shafts (I don't know what the technical word for "shafts" is, but I think you get my meaning). I accepted this as part of the totality of a beautiful picture, but I do agree that it would be better to straighten them, if possible. -- Ikan Kekek (talk) 02:34, 21 January 2016 (UTC)[reply]
  • Have a close look at all the pillars. Not just the nearest ones at the left and right edges of the frame. They are all tapering inwards at the top. Distortion, when it happens due to lens issues, is always strongest at the edges of the frame, but even the columns in the middle of the image (at the back of the room) are more narrow at the top. I'm not going to straighten it when that's how the columns are in reality. Diliff (talk) 07:11, 21 January 2016 (UTC)[reply]
  • Ikan, Pine. The pillars are narrower at the top and this is called entasis. So even the columns in the middle of the picture, which have little perspective/lens distortions show this tapering effect. But those show it fairly symmetrically. As one approaches the sides of the frame, the taper is quite asymmetrical with the edge that is nearer the centre of the photo tapering considerably but the edge towards the outside of the photo tapering very little at all. This is quite pronounced and one doesn't need to view at 100% to see it, but the easiest way to spot it is to open the image in IrfanView or similar and draw a selection rectangle. The verticals that I would expect to be true (window frames, the square bases of the pillars, and the cables for the chandeliers) are fairly true in the image (at 100% there is sometimes a slight angle of a pixel or two, which on a merely 14MP image isn't ideal but not a critical error). I'm not sure of the cause of this asymmetric taper (did you perhaps set vertical control points on a tapering line?) and it may just be the projection.
I know the projection causes the distortion effect one sees at the top of the pillars -- this is where the circular ring at the base of the capital is not parallel with the ground but takes on quite a slope. That's a very common problem with wide-angle rectilinear projections (whether taken with a ultra-wide lens or composed in software) and shows up on pillars and the turrets of buildings. It's a minor enough part of the overall image that I can overlook it here, and isn't something one can fix without artificially Photoshopping the subject like one might do to make a supermodel thinner.
Diliff, unless the columns in this church have an unusual asymmetrical entasis, then the effect seen here is due to the projection/software. I don't know if you can try to minimise the effect but otherwise we'll just have to accept this as a consequence of the wide-angle projection. And in that case, it is a matter of taste whether one finds it too objectionable. -- Colin (talk) 08:19, 21 January 2016 (UTC)[reply]
All perspective distortion is asymmetrical when considering the left and right side of an off-centred object though, because one side is closer to the edge of the frame than the other and the effect of the perspective is greater at the edges. That's not really what Pine is saying though. There is one pillar (the third pillar from the right) that is clearly leaning outwards, but the other pillars are not leaning outwards in the same way, which leads me to conclude that that one pillar is physically leaning outwards. There is another pillar behind and to the left of it that is clearly not on the same vertical axis and is much straighter. Given I am confident that the image does not contain major stitching issues and the lens used does not have any distortion problems, I have to conclude that it is an accurate portrayal of the scene and not caused by any kind of perspective or lens distortion. Let me make some basic claims that I think are fair.
  1. The columns are inconsistent in their angles and feature entasis, meaning they taper inwards at the top and are therefore difficult to use to determine leans or angles. As a minor side issue, they may even be inconsistent when compared to each other due to being difficult to carve the stone with a high level of accuracy at the time.
  2. The window frames and internal lead linings are consistent and most likely truly vertical. The three chandelier chainlinks, acting as plumb lines, are also consistent and truly vertical.
Given the above points, if we can identify that there are consistent vertical lines in the image and the columns do not adhere to these known verticals, I think it's fairly easy to say that the columns are not consistently vertical. I will acknowledge that there is a small overall lean to the right in the image as seen in many of the known verticals, but it is imperceptible except when pixel peeping, and the more obvious perceived rightward lean of some of the columns is an actual physical lean, not an issue with the rotation/distortion of the image. Diliff (talk) 09:41, 21 January 2016 (UTC)[reply]
Confirmed results:
Result: 19 support, 1 oppose, 0 neutral → featured. /Yann (talk) 13:11, 26 January 2016 (UTC)[reply]
This image will be added to the FP gallery: Places/Interiors/Religious buildings
Retrieved from "https://commons.wikimedia.org/w/index.php?title=Commons:Featured_picture_candidates/File:St_Stephen_Walbrook_Church_Interior_1,_London,_UK_-_Diliff.jpg&oldid=185690647"