Barton on Sea...

We love a challenge! If you have an image that you think can be better, post it here and see what the rest of us can do with it.
mikemeister_admin
Posts: 4925
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Wed Apr 13, 2011 2:02 pm

A simple ROC filter at default settings, then a slight contrast increase.  I used the original picture that's just above this post.
Attachments
barton_original-copy-2-jpg
barton_original-copy-2-jpg (122.91 KiB) Viewed 1463 times

mikemeister_admin
Posts: 4925
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Sat Apr 16, 2011 5:10 am

Since I don't know what this scene actually looked like, I thought I'd adjust the yellow slider in the ROC filter to give a warmer picture.  I used the original unaltered pictured (Barton original) that was posted, then used the ROC filter and slid the yellow/blue slider to a -5.
Attachments
barton_original-copy2-jpg
barton_original-copy2-jpg (132.42 KiB) Viewed 1463 times

ggroess
Posts: 5342
Joined: Wed May 24, 2006 2:15 am
Contact:

Postby ggroess » Sat Apr 16, 2011 1:35 pm

I'd say this one looks better art...but I'm not the final judge...Lee is... 8)
He'll be back this week after being off the grid for a week of vacation...or so he told me...

Greg

mikemeister_admin
Posts: 4925
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Sun Apr 17, 2011 12:04 am

I agree Greg, it has a warmer look.

leeharper_admin
Posts: 263
Joined: Tue Aug 20, 2013 2:39 pm

Postby leeharper_admin » Tue Apr 19, 2011 2:46 pm

I'm back... Thanks to all who have participated in this exercise - it has been really interesting - and very helpful - for me to see (and compare) everyone's corrections; my student will be extremely excited to see how you have all improved her image!

One of the aspects that make this image interesting to me is that it was shot on Kodak Portra film. I initially thought that the photographer's choice of film stock ought to be respected - hence the link in my original post to the Flickr Portra group... Offline discussion with Greg has been rattling round my head for a week or so, and I'm now of the opinion that the film stock was not deliberately selected (her photography tutor gave her some free film), and therefore the image might benefit from a correction that doesn't aim too exactly at the 'Portra look'. I know that this contradicts what I've previously said about the image  ::)

I will ask the photographer what she prefers...

In evaluating everyones' corrections I have been playing with some recent advice on the Applied Color Theory Yahoo! group - namely that a blend between multiple corrections will usually look better than any single correction; the final image that I have settled on seems to bear this out - but I think that such blends work best when multiple people have made the corrections (because we all tend to focus on different aspects of the image).

My final version (attached) is based on a average between everyones' corrections, to which I added 15% of Steve's image in Luminosity mode, 25% of Greg G's image in Hue mode; I then slightly boosted the saturation...

I will add a separate post with brief comments regarding the individual posted corrections.

Thanks so much,
Lee.
Attachments
barton-on-sea-challenge-jpg
barton-on-sea-challenge-jpg (245.15 KiB) Viewed 1463 times

leeharper_admin
Posts: 263
Joined: Tue Aug 20, 2013 2:39 pm

Postby leeharper_admin » Tue Apr 19, 2011 3:09 pm

I thought that it might be helpful if I commented briefly on each image submitted. Please take any criticism as subjective and meant completely constructively - and of course if you disagree, I would love to hear your thoughts :)

In comparing the individual corrections to the blended version (see the previous post), I placed each correction above the blended version and compared it using the Normal, Hue, Saturation, Color and Luminosity blend modes (in Photoshop).

Ulf: The luminosity of the correction was too dark (it seems too dark given how sunny the image looks); the saturation is too low; in terms of hue certain features in the image are good, others are not quite right.

Art: The image is too bright overall; Color too green (visually - unfortunately I couldn't use Hue Clocks to confirm this); the saturation of the sky is too low (yet the saturation of the stones in foreground is too high). The saturation of the hill in the foreground is too even (which flattens out the image).

Steve: The luminosity is too harsh/contrasty (though - as I said before - it looks great applied to the blended version at 15%); The Hue is quite good (slightly too magenta); The saturation level is too high...

Greg M: The luminosity of the image is too flat; The color balance is too blue (again I didn't use hue clocks, but visually it looks that way); Hue (certain features are very good, others are way off [see the attached image, which is this version in Hue mode above the blended version]); the saturation is quite uneven (the yellow flowers and some of the greenery are very good - other areas are undersaturated).

Greg G: Again, given how sunny it looks, the luminosity of the image seems too full; the color of the sky and tree are very good; saturation of cliff is too even (this flattens out the image - see the attached butterfly split - but I appreciate that this may have been to mimic Portra); the hue is good apart from the yellow flowers. I decided that I found the hue to be slightly too red, but obviously this was done to make the image look like Portra. I will ask my student which version she likes better...

Again, I am extremely grateful to all of you for posting your corrections - they have all been very helpful! I hope that my comments above are helpful to you all. I'd love to hear your opinions about each version too; I think that hearing each other's opinions is a great way to develop, as we all tend to focus on different aspects of any image that we work on...

Thanks so much,
Lee.
Attachments
barton-on-sea-challenge_h-jpg
barton-on-sea-challenge_h-jpg (131.14 KiB) Viewed 1463 times
barton-on-sea-challenge_s-jpg
barton-on-sea-challenge_s-jpg (114.78 KiB) Viewed 1463 times

sjordan93436
Posts: 460
Joined: Sun Jul 11, 2010 4:23 am
Contact:

Postby sjordan93436 » Tue Apr 19, 2011 11:03 pm

BTW- Lee, my contrast was too strong.  My technique was interesting (to me anyway).  I did an adjustment layer and converted the image to black and white.  I messed with sliders until I got a contrasty black and white.  Then I changed the mode to luminosity.  I probably should have cut the opacity also. 

leeharper_admin
Posts: 263
Joined: Tue Aug 20, 2013 2:39 pm

Postby leeharper_admin » Wed Apr 20, 2011 9:59 am

Steve,

Your technique is interesting to me too! I have actually been experimenting with exactly the same move this week, as a method of compensating for the Helmholtz-Kohlrausch effect. I noticed that you are on the ACT list - did you read Michael Jahn's article about H-K? If you missed it, it's here: https://docs.google.com/View?id=dfwf37zj_1100gnqjnf8h

I think that creating a strong B/W image - and then blending it into the rest of the file using the Luminosity blend mode - is a good way of tackling the tonal portion of a correction; I have found though that when I take that approach I need to cut the opacity of the layer quite significantly. I think that the Helmholtz-Kohlrausch effect may partially explain why the layer opacity needs to be so low: the brightness of objects within an image will change (in the perception of the viewer) according to the objects' saturation level; when creating a Black and White conversion, the contextual cues that the image's saturation level would provide are absent - therefore it is almost impossible to avoid changing the brightness of specific objects in a way that makes no sense in the context of their saturation level.

Applying the B/W conversion in Luminosity mode - at a low opacity - reduces the perceptual disconnect between the tonal adjustment and the existing level of saturation in the image, but only if the opacity is low enough (to make sure that it is I make a habit of increasing the opacity level from 0%, rather than reducing the opacity level from 100% - which I find encourages me to leave it set too high).

Of course, if you only apply the B/W conversion layer at ~10% opacity the question becomes "Does the quality gain justify the time spent creating the adjustment?" I'm inclined to prefer spending my time on other aspects of the correction (in terms of image tonality I tend to spend quite a while hiraloam sharpening, for example)...

As I say though, I am using the same approach as you - I'm just using it for a slightly different effect (and later in the correction - i.e., after I have made my saturation adjustments). Again, if you've not read Michael Jahn's article on the Helmholtz-Kohlrausch effect I would definitely recommend it. There is a bit of math and technical stuff in there that didn't make any sense to me - but also some good pictorial demonstrations of the effect that make the whole thing practical and intuitive.

Again, thanks for the comments (and your correction). Incidentally, have you downloaded everyone's corrections to compare them? It's really instructive (and interesting)...

All best,
Lee.

mikemeister_admin
Posts: 4925
Joined: Fri Sep 20, 2013 8:29 pm

Postby mikemeister_admin » Thu Apr 21, 2011 2:08 am

Lee, it's always good to hear criticisms, since that's how we learn. :)

On thing that I find that makes editing pictures difficult is that we don't know what the scene actually looked like, so we have to use our judgement to come up with a good end result.  I think the end result is heavily influenced by our own tastes, so what looks good to one person, may not look good to another.

For example, in the picture we worked on in this thread, we have no way of knowing what color the sky actually was when the picture was taken.  Since we all feel that it should be some shade of blue, we try to make it look bluer.  If the actual color was a very light blue, then by making it a darker blue, we are forcing a color change in the picture that will affect everything else in the picture.

If the sand actually did have a reddish hue to it, we'd think that was wrong since sand is usually beige, so we'd try to make the sand look beige, and again, change the other colors in the picture.

The obvious way to make these changes would be to place everything on it's own layer and work on it there, that way we wouldn't be forcing all the colors to change into a color that wasn't actually there.

The thing is, when we do that...work on everything in it's own layer, we are re-creating the picture.  Our end result is not what the scene actually looked like, but it does look good to us.  The dilemma is which avenue should we pursue....make it look like the actual scene, or make it look good to us?  I think that most people would opt for making it look better to us, since it's more pleasing to the eye.;)

leeharper_admin
Posts: 263
Joined: Tue Aug 20, 2013 2:39 pm

Postby leeharper_admin » Thu Apr 21, 2011 8:01 pm

Art, I think that you're absolutely right - I think that the goal of any correction ought to be to make the image look good to our audience (even if that just means ourselves); I don't think that trying to make the image look 'correct' is the best approach. Of course making the image 'look correct' involves altering it from what the camera captured - even if the in-camera white balance and exposure are expertly set - as we don't see the world in the same way that cameras record it...

I tend to run into a lot of images where I don't know what colors I should shoot for. Fairly recently I was working on a picture taken at Bryce Canyon; never having been there my original correction was a poor reflection of the location. I am finding that in these situations if I can find some good photographs taken at the location it helps me enormously - it's a way of creating new 'memory colors'. I tend to start at www.fotopedia.com and see what they have, as it covers the whole world and the quality of the photography is very good (so I can place a little more trust in the colors).

In a situation like the Barton-on-Sea image, it's not a well known spot - so an internet search doesn't help much. In these cases you're right that it is tricky not knowing what colors to correct to. This forum is great in these situations, and in this particular case everyones' input has been invaluable to me; in looking at all of the corrections and taking the average I think that we can reach a consensus as to what the colors probably ought to look like.

Obviously looking at the average of all corrections doesn't help whilst you're actually working on a correction; in this case I believe that getting the right colors is more a case of avoiding the wrong ones. This involves a little deduction: if the sun was setting then the sand might have a reddish hue to it - but if the sun was setting there would be a distinctive shadow pattern, etc.

When you mention that enriching the sky (beyond how it actually looked to the photographer) forces a color change into the picture that effects everything else, this really depends on which color space you work in; you could certainly make such a change without penalty if you were working in Lab - HSB would be another option (if the saturation channel was set up differently in CM - i.e., if the horizontal axis defined hue and the vertical axis defined saturation)...

Check out Fotopedia if you've not looked at it before - I think you'll like it  :)

Cheers (and thanks for the comments),
Lee.


Return to “Want us to work on one of your images?”

Who is online

Users browsing this forum: No registered users and 6 guests