Pages: [1] 2   Go Down

Author Topic: A study in blue  (Read 7449 times)

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
A study in blue
« on: July 19, 2011, 10:08:14 am »

...
« Last Edit: October 29, 2011, 10:17:08 am by Justan »
Logged

degrub

  • Guest
Re: A study in blue
« Reply #1 on: July 19, 2011, 03:14:56 pm »

Nice ! For what it is worth , i found the image more appealing when i took out the lower 25 % - the dark sand bar/beach area.
Logged

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #2 on: July 20, 2011, 10:45:54 am »

Thanks for the complement and the intriguing suggestion!

PeterAit

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4559
    • Peter Aitken Photographs
Re: A study in blue
« Reply #3 on: July 20, 2011, 10:50:36 am »

Nice ! For what it is worth , i found the image more appealing when i took out the lower 25 % - the dark sand bar/beach area.

I actually like the sandbar, but would prefer for the tide pool at the very bottom of the image to be either totally or totally out of the image. Having it disappear off the edge is distracting to me.
Logged

francois

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13791
Re: A study in blue
« Reply #4 on: July 20, 2011, 10:56:20 am »

I actually like the sandbar, but would prefer for the tide pool at the very bottom of the image to be either totally or totally out of the image. Having it disappear off the edge is distracting to me.
I agree! At the very least, a bit more of the tide pool at the bottom would be nice.
Logged
Francois

Lonnie Utah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 277
Re: A study in blue
« Reply #5 on: July 20, 2011, 03:54:23 pm »

Level the horizon line... 
Logged

Karl Johnston

  • Newbie
  • *
  • Offline Offline
  • Posts: 8
  • Canadian Adventure Photographer & Travel Writer
    • Website
Re: A study in blue
« Reply #6 on: July 20, 2011, 11:48:24 pm »

I prefer without the sandbar too but it is a great pano.

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #7 on: July 21, 2011, 10:29:15 am »

Thanks for the feedback.

I did an edit to remove the tidal pool & sandbar and another to remove just the tidal pool but didn’t like the result in either case.

I did level the horizon (Thanks for pointing that out). I'm gonna have to start checking that as an early part of the work flow as I frequently miss it.

fike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
  • Hiker Photographer
    • trailpixie.net
Re: A study in blue
« Reply #8 on: July 21, 2011, 10:50:20 am »

people go a little bit crazy about leveling horizons.  Two points to remember:

1) if you can see any land on the other side (say with a lake or bay) the horizon may not look level.  That is okay.
2) The horizon shouldn't be a straight line. unless you have a telephoto, it should appear a bit curved. Trying to make it too straight can make other things in foreground look awkward (piers and pilings for example might end up
Logged
Fike, Trailpixie, or Marc Shaffer

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #9 on: July 21, 2011, 12:04:47 pm »

Hi Marc,

Thanks for the comments. Some people have a very strong sensitivity to what is level. Unfortunately, I'm not one of them. Prior to my fix, the horizon on the right was about ¼ higher than the left, so in this case the cure was a simple rotate around the center and nothing further was needed, but I agree that one can become obsessed with this kind of thing.

The stitching program I use (ptgui) has several optional projections to help solve classic problems related to stitching, but what I just found in PS is the “Guide” tool (under View), which can draw a vertical or horizontal line where desired. I typically use the ruler tool to draw a vertical line based on something in the image and then use transform, but here there is no vertical reference, so the guide tool was perfect for the job.

What is remarkable to me is that both of the bubble levels on my ‘pod were on the mark, yet there was still a slight tilt in the final image.

Slobodan Blagojevic

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 18090
  • When everyone thinks the same, nobody thinks
    • My website
Re: A study in blue
« Reply #10 on: July 21, 2011, 12:50:20 pm »

...
What is remarkable to me is that both of the bubble levels on my ‘pod were on the mark, yet there was still a slight tilt in the final image.


Which both? Where were the two placed?
« Last Edit: July 21, 2011, 01:39:04 pm by Slobodan Blagojevic »
Logged

fike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
  • Hiker Photographer
    • trailpixie.net
Re: A study in blue
« Reply #11 on: July 21, 2011, 01:25:59 pm »

If the pano was captured level, and the center-point of the frame was pointed level at the horizon and the horizon in the processed image file was not horizontal, then you probably had the center (yaw) and/or the horizon (pitch) of the image set wrong in PTGui.  I don't know how much monkeying around you do with the projection and the center point in the panorama editor, but if you do that wrong, most specifically setting the center point to the left or right of center, you can mess up a level horizon.  More typically, you have the camera pointed slightly downward, and this requires some intervention in PTGui. 

Are you using a spherical or a cylindrical head?
Logged
Fike, Trailpixie, or Marc Shaffer

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #12 on: July 22, 2011, 09:33:46 am »

Which both? Where were the two placed?

There is one bubble level on the tripod itself, and the head has I think 4 or 5 of them, so that one level is useful no matter which position is being used. The ‘pod is typically very reliable. One side probably sunk a little in the sand after I did the initial setup.

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #13 on: July 22, 2011, 09:35:47 am »

If the pano was captured level, and the center-point of the frame was pointed level at the horizon and the horizon in the processed image file was not horizontal, then you probably had the center (yaw) and/or the horizon (pitch) of the image set wrong in PTGui.  I don't know how much monkeying around you do with the projection and the center point in the panorama editor, but if you do that wrong, most specifically setting the center point to the left or right of center, you can mess up a level horizon.  More typically, you have the camera pointed slightly downward, and this requires some intervention in PTGui. 

Interesting. I’ll have to check. I don’t typically change PTGui from the defaults.

Quote
Are you using a spherical or a cylindrical head?

Sorry but I don’t know how those terms apply to tripod heads.

fike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
  • Hiker Photographer
    • trailpixie.net
Re: A study in blue
« Reply #14 on: July 22, 2011, 10:36:29 am »

PTGUI doesn't always get things perfectly correct.  Sometimes you end up using the panorama editor to drag the image around and rotate it.  It sounds like you are using it in a fairly automatic mode.

One thing you can do in PTGUI to make sure the horizon is straight is manually enter control points and specify horizontal lines in the control point. With this control point, you set a single point along a straight line, then on an adjacent image, you specify another point (not the same one) along the same line. If you enter a few of these along the horizon, the auto optimizing will make that line horizontal...sometimes to the detriment of vertical lines or other horizontal lines...but that is unavoidable.

If you don't know which pano head you are using, I am assuming you are using a standard ballhead of some sort and then rotating the camera with that. For this image, that is generally fine. Purpose built spherical and cylindrical heads are not really needed for panos where the subject is far away.

You mentioned that everything was level.  Was the camera platform level throughout the range of panning?  That really is the only one that matters.  The problem with ballheads for pano work is that the platform may be "close enough" to level, but when you pan the platform around the base of the ballhead a small error can be compounded to a much larger error.  When you are leveling a camera for using the ballhead panning adjustment, you should probably get a bubble level for the hotshoe bubble level and check it for level along the entire panning range you intend to use.  This prevents you from being in a situation where the tripod legs are mostly level and you use the ballhead to make the camera perfectly level, but as you pan the camera is no longer level because you are rotating around the base of the ballhead.  This can be avoided by either putting a panning clamp on top of the ballhead or by putting a leveling base under the ballhead.  the cheapest solution is the hotshoe bubble level and careful leveling of the legs.

http://www.bhphotovideo.com/c/product/419395-REG/Jobu_Design_LVL_PL_2_Axis_Flash_Hot.html
Logged
Fike, Trailpixie, or Marc Shaffer

John R

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5248
Re: A study in blue
« Reply #15 on: July 23, 2011, 08:14:24 pm »

I read somewhere that the new in-camera levels are not necessarily accurate. But of course you are referring to external levels. I have been to many seminars by professionals in the last 20 years, and most care little about horizontal straighness unless it is obvious or detrimental to the image. Moreover, most freely admit to pointing their cameras up or down to get the proportions of sky and land they want and do there best to make the image image level within that context.

As for the image itself, I have to say, I find the bottom sandbar too powerful and dominant, to the point that, IMO, it ruins the image. It also grabs the attention of the viewer and immediately the eye out of the image. IMO, the black line or sandbar in the middle succeeds because it is slender and does not upset the balance of the rest of the image.
Logged

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #16 on: July 25, 2011, 11:44:05 am »

> PTGUI doesn't always get things perfectly correct.  Sometimes you end up using the panorama editor to drag the image around and rotate it.  It sounds like you are using it in a fairly automatic mode.

One thing you can do in PTGUI to make sure the horizon is straight is manually enter control points and specify horizontal lines in the control point. With this control point, you set a single point along a straight line, then on an adjacent image, you specify another point (not the same one) along the same line. If you enter a few of these along the horizon, the auto optimizing will make that line horizontal...sometimes to the detriment of vertical lines or other horizontal lines...but that is unavoidable.

Thanks for the tips! I always appreciate this kind of information! I’ll spend some time playing with control points to affect the horizon next time I have the opportunity. I do have a question about PTGUI, I thought to impart a slight curve into the horizon line, to vaguely simulate the curvature of the earth. I declined to peruse it because the range of focus wasn’t wide enough to justify this kind of gimmick, but how would one do that in PTGUI?

> You mentioned that everything was level.  Was the camera platform level throughout the range of panning?  That really is the only one that matters.

It was when I set it up. The tripod has one bubble level and the head has 5. This provides for 2 a bubble levels that enable an accurate read. I'm guessing that one of the legs sunk a little in the sand.

At some point I’ll get an actual pano head. The RRS brand is the leader. Not too long ago Bill T posted an article about a home brew pano head he made. I’d love to fabricate one following the one he made, but I haven't’ taken the time. So far I have avoided the problems associated with shooting panos of nearby subjects and other places where the utility of a pano head shines through.

Justan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1928
    • Justan-Elk.com
Re: A study in blue
« Reply #17 on: July 25, 2011, 11:44:47 am »

> I have been to many seminars by professionals in the last 20 years, and most care little about horizontal straighness unless it is obvious or detrimental to the image. Moreover, most freely admit to pointing their cameras up or down to get the proportions of sky and land they want and do there best to make the image image level within that context.

In many cases this works fine, but because some have a sensitivity for this kind of thing….

> As for the image itself, I have to say, I find the bottom sandbar too powerful and dominant, to the point that, IMO, it ruins the image. It also grabs the attention of the viewer and immediately the eye out of the image. IMO, the black line or sandbar in the middle succeeds because it is slender and does not upset the balance of the rest of the image

Interesting. What would you do to change the composition?

John R

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5248
Re: A study in blue
« Reply #18 on: July 25, 2011, 01:02:56 pm »


Interesting. What would you do to change the composition?

I have to admit on second look it is not as bad as I first thought. But I cannot improve on the image as is, only suggest that more foreground could have been included under the bottom sandbar and that would help diminish the power of the sandbar and make it more like the top sandbar- a nice horizontal visual RESTING point as my eye explores the rest of the image. But then again, this might be too subjective on my part and this image when seen large could be fine.
Logged

fike

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1413
  • Hiker Photographer
    • trailpixie.net
Re: A study in blue
« Reply #19 on: July 25, 2011, 01:48:02 pm »

I do have a question about PTGUI, I thought to impart a slight curve into the horizon line, to vaguely simulate the curvature of the earth. I declined to peruse it because the range of focus wasn’t wide enough to justify this kind of gimmick, but how would one do that in PTGUI?

Using the Panoramic Editor window (the one where you can drag the image around, crop the image, and change the projection) you will want to drag the image upward, effectively moving the horizon below it's actual position.  You could also do this with numerical transform (I forget the exact menu title) and applying a slightly negative number to the pitch value.
Logged
Fike, Trailpixie, or Marc Shaffer
Pages: [1] 2   Go Up