WaxPhoto
I AM SAM
Premium
join:2004-04-08
Roanoke, IN
| Punch Up Those Photos (Gimp Style)
Canon EOS 10D
17mm 1/180th F8 ISO100
ET's original | 
GIMP LAB Overlay Method | 
Canon EOS 10D
17mm 1/180th F8 ISO100
ET's apllication of Photoshop LAB Overlay Method |
 posted here: »Punch up those photos.......fast and easy.
A note before we begin: Its important to keep in mind Gimp treats the LAB colorspace very differently from Photoshop. For one, any changes you make will not be be visible until you do the final step. Gimp decomposes an image to the LAB colorspace using Grayscale layers, as opposed to PS's use of channels. In practice, this is more convenient for some operations, but less convenient overall. For this method... it is a bit more of a pain in the ass, with more steps. Here are those steps:
1)Open Image
2)Go to Image>Mode>Decompose
3)Check LAB, Check Decompose to Layers, hit OK
4)Open Layers dialog. Make only the B layer visible.
5)Duplicate B layer, set "B Copy" to overlay (Opacity is 100%).
6)Merge both B layers (Image>Merge Visible Layers/CTRL-M) Choose Expand as necessary (though other options don't make a difference in this case) Hit Ok.
7)Repeat steps #4 through #6 for the "A" layer.
8)Select "L" layer, apply a slight S curve, mostly adjusted towards the dark end of the curve. (GIMP's overlay is calculated a bit differently)
9)Image>Mode>Recompose and Make sure you select LAB with the correct layer order.
Thats it, the output will still be slightly more even of an exposure compared with PS... not quite as dark. (This may be a 8/16 bit side effect)
--
*Formerly known as LitteredMind* To see what is in front of one's nose needs a constant struggle. -George Orwell |
|
Edit This
Premium,MVM
join:2001-05-08 | Absolutely phenomenal work WaxPhoto . The Gimp image looks the best to my eyes. |
|
WaxPhoto
I AM SAM
Premium
join:2004-04-08
Roanoke, IN | Thanks... Did you do anything other than what you posted to 3b.jpg? Maybe blurring A&B channels a little? |
|
Edit This
Premium,MVM
join:2001-05-08
 ·Comcast
| said by WaxPhoto :Thanks... Did you do anything other than what you posted to 3b.jpg? Maybe blurring A&B channels a little? As far as the original image....No. I used just the steps I posted. I used two separate layers to create this one....but....I did use the original LAB technique although each layer was processed differently and then combined into the final image. »Re: Punch up those photos.......fast and easy.
--
I post therefore I am. |
|
WaxPhoto
I AM SAM
Premium
join:2004-04-08
Roanoke, IN
| Ok. One last question... In CS2, does this method work in 16 bit mode, and if so, was your output in 16 bit? (for the Image from your first post in that thread)
--
*Formerly known as LitteredMind* To see what is in front of one's nose needs a constant struggle. -George Orwell |
|
34140721
Small Block Chevys Never Die..
Premium
join:2003-03-25
| reply to Edit This
said by Edit This :Absolutely phenomenal work WaxPhoto . The Gimp image looks the best to my eyes. I concur...the Gimp photo is much better... Many thanks for sharing this and for your outstanding work...  |
|
Andrew J
Premium
join:2001-11-09
Lancaster, PA
clubs:
·Comcast
 ·Vonage
 ·Verizon Online DSL
| reply to WaxPhoto
You seem to be doing a slight variation of this adjustment that takes 3 seconds in Elements and 16 bit:
»www.cambridgeincolour.com/tutori···ment.htm
--
Best Team. |
|
WaxPhoto
I AM SAM
Premium
join:2004-04-08
Roanoke, IN
| This does so much more than local contrast enhancement, though. Disregarding the color change, it may have a somewhat similar effect, but this method is for more than "stretching out" the color DR, which is all LCE really does. This method actually differentiates between hues. Read through some of ET's past LAB links.
--
*Formerly known as LitteredMind* To see what is in front of one's nose needs a constant struggle. -George Orwell |
|
Edit This
Premium,MVM
join:2001-05-08
·Comcast
edit:
January 5th, @01:24PM
| reply to Andrew J
From the link you provided Andrew J :
"Local contrast enhancement, as with sharpening, can also create unwanted color changes if performed on all three color channels. In addition, Local contrast enhancement can increase color saturation significantly. You can eliminate these unwanted effects by either performing local contrast enhancement in the lightness channel of the LAB color space, or in a separate layer (while still in an RGB working space) and blending using "luminosity" in the layers window.
Local contrast enhancement can also clip highlights in regions which are both very bright and adjacent to a darker region. For this reason, it should be performed before adjusting levels (if levels are used bring tones to the extreme highlights within the image histogram). This allows for a "buffer zone" when local contrast enhancement extends the lightest and darkest tones to full white or black, respectively.
Care should also be taken when using this technique because it can detract from the "smoothness" of tones within your image-- thereby changing its mood. Portrait photography is one area where one should be particularly cautious with this technique".
Nice link Andrew J . |
|
Edit This
Premium,MVM
join:2001-05-08
·Comcast
| reply to WaxPhoto
said by WaxPhoto :Ok. One last question... In CS2, does this method work in 16 bit mode, and if so, was your output in 16 bit? (for the Image from your first post in that thread) Yes it works in 16 bit mode.......No my posted output was 8 bit not 16.
--
I post therefore I am. |
|
WaxPhoto
I AM SAM
Premium
join:2004-04-08
Roanoke, IN
| said by Edit This :said by WaxPhoto :Ok. One last question... In CS2, does this method work in 16 bit mode, and if so, was your output in 16 bit? (for the Image from your first post in that thread) Yes it works in 16 bit mode.......No my posted output was 8 bit not 16. Ok. So now all I have to determine is if GIMP uses a different algorithm for overlay blending mode, compared to PS. If it doesn't, then it must be the LAB process that is different.
--
*Formerly known as LitteredMind* To see what is in front of one's nose needs a constant struggle. -George Orwell |
|
timcuth
Braves Fan
Premium
join:2000-09-18
Pelham, AL
clubs:
·AT&T Southeast
edit:
January 16th, @07:03PM
| reply to WaxPhoto
Canon PowerShot A60
13.4mm 1/640th F4.5
|  |  |
Here are three samples. Unedited is the shot straight from the camera. lab_no_contrast is only doing the overlay technique on the decomposed B and A layers. lab_contrast adds the s-curve to the L-layer. Does it appear that I am doing things, correctly?
I have more questions, but I will save them for a while.
Tim
--
I'm up on a tightrope. One side's fear and one is hope. - Leon Russell |
|
timcuth
Braves Fan
Premium
join:2000-09-18
Pelham, AL
clubs:
·AT&T Southeast
edit:
January 16th, @07:04PM
| reply to WaxPhoto
PS - Sorry, I should have done something to make the files smaller. I was so focused on the technique I neglected to do so.
PPS - Never mind.  I figured out how to fix it.
Tim
--
I'm up on a tightrope. One side's fear and one is hope. - Leon Russell |
|
bobny1
join:2004-09-10
Bronx, NY
|  | 
Canon EOS DIGITAL REBEL XT
ISOARRAY(0xb666f60)
|
|
|
