June 2008 Contest Discussion

[Maggie42]

I thought we could not change them once posted. I have a couple of mistakes on mine but will not change it.

I agree about the quality, it's going to be really hard to choose. I love them all!

[LonK]

Maggie42, from Contest FAQS thread:

--------------
Can I change my entry?

You can change your entry until the end of the contest.

Note you are not allowed more than one entry so you will have to replace your original entry or delete it. If there are more than one entry for a person at the end of a contest, all entries for that person will be disqualified.
--------------

[T Paul]

Here is a link to a temporary album for easier viewing: June Content Entries

[LonK]

Originally Posted by Kraellin

...(i'd have put you higher Lonk, but it was so damn good it had a painting quality to it instead of a photo and again, the rules stated "realistically". i've never voted someone down for being too damn good before.)
...
also, i'm curious in general; how many saw leggings/tights on the women's legs?

I'll accept that as a compliment, Craig. However, for all intents and purposes, colorization IS painting

As far as "leggings/tights", I believe it would have been considered sacrilegious for a women to expose her legs in public at the turn of the century. This is a ballet scene, not burlesque, so I think it would be likely (even though it is not apparent in the B&W photo).

[asims28]

I have a quick question. I am on a different computer at work and i realized that my entry looks ALOT darker than on my computer at home... i dont know if its just this computer or what. Is anyone else seeing this?

[dweekley]

asims, would you mind sharing some info (or maybe someone else with more color knowledge can help too) on how you got the saturated wood color on your entry? it looks like a walnut to me and i wasn't able to get that although i tried. questions that come to mind....what mode did you work in, what 'color' is it, did you adjust the orig image in anyway to influence this, what layer type did you work on?

as for your image being overly dark, i personally didn't see that on my computer. i liked your entry it was just more saturated than what i was looking for in the results.

Doreen

[LonK]

I felt your entry was a bit dark and contrasty, asims28. Of course, that's a subjective observation. Looking at the histogram, I see considerable shadow clipping. For a quick check of your monitor(s) brightness, look at the header of any page here at RetouchPRO.com. You should see 10 equally sized, distinct greyscale blocks ranging from white to black If you don't (e.g. the darker blocks are indiscernible), you need to calibrate.

[asims28]

dweekley-
I started with an orange color so i could get a gold chair, but didn't like it and turned it brown. I may not be exact with the numbers but i should be pretty close (it may be too many steps than needed but its exactly how i achieved the color), I went back into the original image and tried to duplicate my color.
I first copied desaturated layer and used adjustment- hue/saturation- colorize.
hue 22; sat 66; lightness -31
Tweaked the color balance
-38 cyan; -24; mageta
Then my favorite "cheat" adjustment-variations with mid tones selected
more cyan; darker
Tweaked more in color balance
-38 magenta
Turned down saturation a bit
Finally i turned down the contrast so it wouldn't look so shiny around -70
The way i colored the entire picture is with layer masks, it took forever.

LonK-
Thanks for the info, im back at home and it looks fine on this computer, i can tell the color differance of each grey block.. i agree it was a little too much contrast, i was about to lighten it up a bit and realized the contest was already closed.

[Ruzzez]

The light on the legs of the girl that is standing seems too bright without enough contours to me not to be leggings. It is not so obvious for the girl sitting.

I thought that leggings were more appropriate.

While we are questioning the choices. Does anyone else feel that the eye of the girl sitting was whitened by a few to the point of making her look cross eyed?

[dweekley]

asmis28-
thanks ahead of time, i'll give that a try.

ruzzez-
balletKGM was in my top 8 but ended up cut due primarily to the sitting girls eye and a spot or two of missed coloring.

[Kraellin]

I'll accept that as a compliment, Craig. However, for all intents and purposes, colorization IS painting

lonk, indeed, it is. bear in mind that i'm viewing this as a restorer. and to me, restoring is 'bringing back to the original condition'. what i saw in yours was BETTER than original condition and whereas that is desirable to some, as a restorer, i dont find it 'realistic'. and that's why i didnt give you first place in this. of course, it could well be argued that since there was no color photography at all when the picture was taken, that there is no way that any colorizing could be 'realisitic' in the truest sense, but, there were colorization artists in those days, so again, i felt your work was BETTER than those works would have been in that day and time. i do hope that makes sense.

[LonK]

Originally Posted by Kraellin

...i'm viewing this as a restorer. and to me, restoring is 'bringing back to the original condition'.

It is, and I could agree more readily if the task for the contest was: Restore this photo as if it were originally a hand-colored print. In that case, "realistically colorize" might dictate a different approach. There was, however, no mention of definitive restoration and/or period colorization. Then again, the typical sketchy task basis and latitude for interpretation of RetouchPRO contests contributes to the enjoyment. Kudos to our imaginative contest creators for keeping us on our toes!

... of course, it could well be argued that since there was no color photography at all when the picture was taken, that there is no way that any colorizing could be 'realisitic' in the truest sense, but, there were colorization artists in those days...

Decades ago I tried my hand at coloring prints with both oils and Prismacolor pencils. Believe me, those results couldn't hold a candle to what can be done these days, digitally -- in terms of realism -- especially since the colorist had little to no control over the quality of the base print. If the print had not been specifically processed with the intention of subsequent manual colorizing, the result would inevitably be unpersuasive. This coincides with my key guideline for digital colorization: Proper preparation of the base image.

Let me also defer to the works of my digital colorization muse, Vikki Hansen, a true maven of the art. For the most part, her results look like pristine, natural color photos that were taken today -- no matter how old or worn the source photograph.

I think another valid test of realistic colorization is to look at a result, without knowledge of the source, and be convinced that it was a genuine color photo from the get-go.

All philosophical positioning aside, I'd just like to say how interesting and beneficial it is to see and get to understand some of the diverse individual takes revealed in these wonderful contests. Each offers a wealth of insight into the craft we all aspire to excel at.

[dweekley]

while i agree that part of the fun (and for me learning experience) is to see how different people interpret the contest instructions, given what i've seen here it looks like it would also be fun to do a colorization that clearly states the goal is to produce a result that doesn't look hand colored but instead looks like it was captured today, via color, camera capture :-)
Doreen

[LonK]

You mean like THIS contest?

*Realistically colorize the provided photo.



Seriously though, one of the best techniques for learning colorization is to start with a full color photo that has been converted to B&W. Then without referring to the color photo (perhaps just some notes on the colors), digitally colorize it. When you think you've done it, compare it side-by-side with the original color photo and see how well you did. I think you'll be amazed at how much you can learn with this exercise.

[Kraellin]

lonk, i simply took the photo as if we were there and then in that time and what colorization would or could have been done in that time period. that's my interpretation of 'realistic colorization' here. that's all. yours didnt meet that criteria the best in my estimation. and that's what i meant by it being too good. it wouldnt have been that saturated (and since you mentioned vikki's work, do look at some of her gallery work and you'll see what i mean ). you also did something odd with the lighting i didnt feel fit the bill... that odd sort of shadowed vignetting. altogether, that's why you didnt get my first place vote.