Wikipedia:Featured picture candidates/Church of St Nicholas and St Mary, Stowey

Add links
From Wikipedia, the free encyclopedia

Church of St Nicholas and St Mary, Stowey[edit]

Voting period is over. Please don't add any new votes. Voting period ends on 1 Apr 2014 at 15:17:32 (UTC)

Original – The sandstone Church of St Nicholas and St Mary in Stowey, Somerset, England
Alt 1 - Tower straightened
Tower straightened, but maintaining some vertical perspective
Reason
Demonstrates the mixed stone architecture of the church
Articles in which this image appears
Church of St Nicholas and St Mary, Stowey
FP category for this image
Wikipedia:Featured pictures/Places/Architecture
Creator
Rodw
  • Support as nominator --— Rod talk 15:17, 22 March 2014 (UTC)[reply]
  • Comment - Tower seems to be leaning inward. Is there lens distortion here? — Crisco 1492 (talk) 15:25, 22 March 2014 (UTC)[reply]
  • I have straightened the tower in Alt 1 - is that better?— Rod talk 16:29, 22 March 2014 (UTC)[reply]
Rod , it appears you may have done a simple rotation but a perspective correction is what is really required. Otherwise you end up with a straight tower and sloped walls (or vice versa). Saffron Blaze (talk) 17:56, 22 March 2014 (UTC)[reply]
Agree that perspective appears distorted, which is distracting & detracts from pic. Sca (talk) 20:37, 22 March 2014 (UTC)[reply]
Do be careful about this. The number of old buildings I've seen that genuinely have a tilt to them is fairly high. Though this does seem far more than could reasonably be expected in this case. Adam Cuerden (talk) 20:52, 22 March 2014 (UTC)[reply]
I've looked up Perspective distortion but I'm still unsure what you want me to do.— Rod talk 21:26, 22 March 2014 (UTC)[reply]
  • Comment - I have uploaded a version with the lens distortion removed over Rod's straightened file (the ALT). Owing to a very close crop, I also had to add some sky. Support ALT1. — Crisco 1492 (talk) 01:11, 23 March 2014 (UTC)[reply]
  • Comment - as usual, the straightening has been overdone. If the tower at one end of the building is made parallel to the wall at the opposite end (i.e. all verticals are parallel) then the end result is that the tower and the chancel wall (the one on the right) are leaning outwards from each other. Combining this with the lie of the land, and we now have a nice little church with a very serious problem of subsidence. Amandajm (talk) 07:51, 23 March 2014 (UTC)[reply]
The problem is that what is referred to on these pages as "camera distortion" is in fact about 90% the effect of visual perspective. Rod's eye level was on the height of the lower edge of the left-hand window (horizontally) and near the centre drainpipe (vertically). This means that every vertical above that window sill was sloping in towards the centre, regardless of whether it was perceived by the human eye of Rod, or by his camera lens. That inwards lean affects the tower in particular, because it is tall. But if you remove the perspective entirely, the building falls apart down the middle.
My adjustment is lower resolution and could to be improved. I have put it here to indicate that a more subtle approach than making the lines parallel would be better. Please feel free to upload a better version over the top. Amandajm (talk) 08:25, 23 March 2014 (UTC)[reply]
I think Crisco misspoke in calling it lens distortion (although as you say, there may have been some of that also), because the distortion that we are talking about here is predominantly perspective distortion and has nothing to do with the lens but the consequences of rectilinear projection of a curved scene onto a flat surface. The human eye does not see the world this way, and this is why there is a fundamental problem between comparing what a camera sees and what the human eye sees. But as I've said many times already, I really don't think the human eye (and its visual perception system) actually sees the inward sloping vertical lines as a rectilinear lens does when tilted upwards. There are many reasons for this, but the main reason is simple: We only see very small sections of a scene at any one time with the centre of our vision, and piece them in our brains to form a cohesive perception of a scene. When our eyes scan around the scene, they centre the view on those vertical lines, and the lines no longer slope inwards as they would if they were off-centred. As such, I honestly believe you are mistaking what our eyes see with what a camera usually captures when you argue for the preservation of inward leaning verticals and this affects your judgement on what photo should look like in order to replicate our perception of a scene.
However, in spite of all of the above, I think you are right about one thing. The tower is subsiding and is leaning in reality. This is evident because although the tower is now vertically corrected in Crisco's edit, the wall and drainpipe in the middle of the building is now leaning considerably outwards. Although this wall too could be affected by subsistence so I'm not sure there are any vertical lines that we can be certain of. Ðiliff «» (Talk) 09:20, 23 March 2014 (UTC)[reply]
  • I agree, I likely misspoke earlier. Rod would probably have to provide feedback regarding exactly how far out everything is leaning, or in, as he's been on-site. I'm a wee bit far away. — Crisco 1492 (talk) 09:51, 23 March 2014 (UTC)[reply]
  • I was thinking of going back and taking more shots of the church (and some interiors) but around it is private land & I'm not sure where I will be able to get (it is also raining here at present). The tower didn't look to have much of a lean unlike This church.— Rod talk 10:04, 23 March 2014 (UTC)[reply]
Amanda, your edit looks somewhat nice, but it needs some more sky (it's a little off-centre). Do you mind if I add it? — Crisco 1492 (talk) 09:52, 23 March 2014 (UTC)[reply]

Not Promoted --Armbrust The Homunculus 15:17, 1 April 2014 (UTC)[reply]

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Featured_picture_candidates/Church_of_St_Nicholas_and_St_Mary,_Stowey&oldid=1138496072"