RetouchPRO

Go Back   RetouchPRO > Technique > Photo Retouching
Register Blogs FAQ Site Nav Search Today's Posts Mark Forums Read


Photo Retouching "Improving" photos, post-production, correction, etc.

Automatic Skin Smoothing.

Reply
 
Thread Tools
  #81  
Old 11-01-2010, 11:08 PM
ShadowLight ShadowLight is offline
Senior Member
 
Join Date: Oct 2009
Posts: 397
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by MikeyPee View Post
First off lemme just say this plug-in is fantastic. I was a little confused by the way it worked at first but I'm starting to get the hang of it. There are a few things I had questions about however:

- What is the benefit to using the "Exact" and "Variable" pore adjustment bar?

- I was testing the plug-in on one of my photos and I saw absolutely no benefit when moving the bar towards the "Blotches" side. In fact, it made the image significantly worse the more a shifted it to that side. It actually would add blotchiness to the photos rather than take it away.
I'm glad you like it. Exact-Variable slider give you the option to control how much of variation of the "blotches" or "pores" to correct at the same time.
the more targeted correction you make, the better result you will have. (don't overuse it when it is not needed)

The program makes calculation on the maximum size of blotches you may need, however it takes account of the image size in pixels, it does not detect the face of the portrait and how many pixels it is, so some values towards the end of "blotches" may give you negative effect (make it worse), the same with the "Strength" you use. In some cases using more than 50% will be beneficial, but if you push it too much it will make the image worse.

So play with the sliders until you get the hang of it, and things will get easier from there.
Reply With Quote top
  #82  
Old 11-02-2010, 12:04 AM
bakerser bakerser is offline
Member
 
Join Date: Jan 2008
Posts: 91
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by ShadowLight View Post
I pointed out that the even the evangelist of hand-separation Sean uses HP when it comes to the technique I'm using.
Correct. Because in this post I elucidated how the manual technique is identical to using HP on a -50 Contrast copy of the image (with either B/C version).

Quote:
Originally Posted by ShadowLight View Post
Personally I can't visually see a difference other when testing the full band (when it comes to adjustment layer), however when contrast is applied directly it does shift the lum. as you can see clearly in the tests. (or do we have to close our eye for the sake of using "contrast -50" ?)
We agree that the different versions of B/C give different mean brightness values. What you're still failing to recognize is that the HP of both versions is identical to each other, but markedly different from that of a 'naive' HP on a full-contrast image (where lots of detail is clipped to 0 or 255). Again, this is something which I've explained at length both on MM and on my blog. The information is free and I really encourage you to take the time to finally read it. There's no free lunch, and you just can't map 512 potential values down to 256 without losing something. You can lose 128 on each end (losing highlight and shadow detail), or such up a bit of rounding error which will have less impact than NR'd sensor noise anyway. You wouldn't be having this conversation with me if highend users didn't prefer the latter option.

Quote:
Originally Posted by ShadowLight View Post
As I said, show me example why the other way is better, and I will gladly implement it.
So far you haven't done it.
As above.

Quote:
Originally Posted by ShadowLight View Post
Edit:... I'm not sure what you actually ask here... you ask me to use "contrast -50" something that you have "submitted a bug for", and you showed no proof that it is better?
I submitted a report to Adobe that they have a discrepancy in their software which was clearly an oversight in the development and testing process. Two identically named tools should give identical results. What I'm asking you to look at is that the result of FS on any two such 'different' images is the same, because FS doesn't care about what the mean brightness of an image started at; it only cares about the relative brightness of pixels over the distance between them. [Now, for the purists & hardcore geeks: yes, this breaks down a little bit in some cases because PS's Gaussian Blur (and by consequence its High Pass) isn't really a Gaussian Blur, but only an approximation of one. That doesn't undo the realities of 512->256, however.]
Quote:
Originally Posted by ShadowLight View Post
or you are just trolling...eh? Sean
Simply clearing the air about what I have and have not said.
Quote:
Originally Posted by ShadowLight View Post
besides he/you didn't invent the frequency separation, so I don't get what the big deals is... what he does is nothing new.
Who claimed that I did? Da Vinci likely had many of the same ideas as we're using (albeit without a modern ability to operate with them) - they're not new.

Last edited by bakerser; 11-02-2010 at 12:12 AM. Reason: Typo!
Reply With Quote top
  #83  
Old 11-02-2010, 12:36 AM
ShadowLight ShadowLight is offline
Senior Member
 
Join Date: Oct 2009
Posts: 397
Re: Automatic Skin Smoothing.

So you are clarifying that you have misunderstood the context of what was said... ?

let me clarify it again
some "expert" oneredpanter said that s/he likes the application, but if only it used "apply" for higher precision, s/he would use it all the time.

then I pointed out that you use HP when you use what you call invert-high-pass, and that it turned out that I use alpha, and you use contrast -50, to get what each is after.

I had to point out that you have failed for several years to notice that your way does not even get you 127-128, but has a shift of the values.
The HP is the same, but the Contrast -50 correction will have arbitrary shift in lum.
just stating the facts.

if you like to show an example how your way is better, please do
(make a psd and we can compare it)

otherwise there isn't a point about discussing it more
take care
Reply With Quote top
  #84  
Old 11-02-2010, 04:54 AM
MikeyPee MikeyPee is offline
Junior Member
 
Join Date: Sep 2010
Posts: 5
Re: Automatic Skin Smoothing.

thanks for the clarification shadowlight
Reply With Quote top
  #85  
Old 11-02-2010, 05:02 AM
bakerser bakerser is offline
Member
 
Join Date: Jan 2008
Posts: 91
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by ShadowLight View Post
I had to point out that you have failed for several years to notice that your way does not even get you 127-128, but has a shift of the values.
You keep coming back to this, but I don't understand why you feel that I should have paid attention to something which is completely inconsequential?

Quote:
Originally Posted by ShadowLight View Post
if you like to show an example how your way is better, please do
(make a psd and we can compare it)

otherwise there isn't a point about discussing it more
And we went through that over at MM, but you refused to acknowledge it when presented to you. I can't help that.

I implored you there and I'll implore you again here now - look at who is agreeing with me on methodology and ask yourself whether you are really the only one among all these brilliant folks who knows the "truth" about how to do proper frequency operations in PS, or whether we might collectively know a thing or two about it, having written more on the topic in the past couple years than had existed in the entire history of PS before then.
Quote:
Originally Posted by ShadowLight View Post
The HP is the same
At least we've gotten that through to you.
Quote:
Originally Posted by ShadowLight View Post
but the Contrast -50 correction will have arbitrary shift in lum.
The shift is not arbitrary. It can be predicted exactly what value PS will give. And, as you've acknowledged that the HP is the same (just stating the facts), is ... completely inconsequential.
Reply With Quote top
  #86  
Old 11-02-2010, 07:00 AM
ShadowLight ShadowLight is offline
Senior Member
 
Join Date: Oct 2009
Posts: 397
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by bakerser View Post
The shift is not arbitrary. It can be predicted exactly what value PS will give. And, as you've acknowledged that the HP is the same (just stating the facts), is ... completely inconsequential.
The shift depends on the overall image content which is unpredictable - and in this light I call it "arbitrary".

With contrast -50, the correction does not result in mid-grey, which should be your main test to check if your concept is working.

In MM we spend a few days with me testing your technique, hitting the shift error, and you insisting that you are right and I'm still wrong and that "something is wrong with my system". When finally a 3rd party confirmed the results I was receiving, you start saying that this doesn't matter.
I do not intend to repeat this fruitless discussion.


Please stop confusing people,
intentionally or not this is not helpful at all.

I don't intend to bother everyone with why it works or the DSP behind it.

if you like to discuss the "why" and you have concerns that the current implementation does not "work" (although you can actually test it and see for yourself it does), please send me a private message and I will answer your questions or concerns there, if you have any.


Quote:
Originally Posted by MikeyPee View Post
thanks for the clarification shadowlight
cheers

Last edited by ShadowLight; 11-02-2010 at 07:45 AM.
Reply With Quote top
  #87  
Old 11-02-2010, 07:51 AM
bakerser bakerser is offline
Member
 
Join Date: Jan 2008
Posts: 91
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by ShadowLight View Post
The shift depends on the overall image content which is unpredictable - and in this light I call it "arbitrary".
The "shift" (which isn't a shift - moving towards 127.5 is the "shift") is based on the input image mean. I'll post a formula for you if it'll help you.
Quote:
Originally Posted by ShadowLight View Post
With contrast -50, the correction does not result in mid-grey, which should be your main test to check if your concept is working.
The correction actually should not result in a mean of middle gray, as it means that the tool has shifted the image brightness when all we asked it to do was to change the contrast. Think about it - if the tool is really only working on the contrast, why does the mean shift? If you know 1/10 as much about math as you're suggesting, you know that shouldn't be.
Quote:
Originally Posted by ShadowLight View Post
In MM we spend a few days with me testing your technique, hitting the shift error, and you insisting that you are right and I'm still wrong and that "something is wrong with my system". When finally a 3rd party confirmed the results I was receiving.
You're right, we both wasted a lot of time on the assumption that you were doing something which made sense. On your end because you hadn't thought through the above; on my end because I thought you couldn't possibly mean what you were actually typing and so was looking for some other problem.
Quote:
Originally Posted by ShadowLight View Post
Please stop confusing people,
intentionally or not this is not helpful at all.
I rejoined this conversation to keep you from doing just that. Maybe you really don't mean to be throwing people off the track towards good retouching - I certainly hope not; you're too good a Flash coder to want to think that.
Quote:
Originally Posted by ShadowLight View Post
I don't intend to bother everyone with why it works or the DSP behind it.
Funny, because I've put quite a number of hours into explaining it to as great a depth as people want to learn it.
Reply With Quote top
  #88  
Old 11-02-2010, 08:43 AM
Flashtones Flashtones is online now
Senior Member
 
Join Date: Apr 2009
Posts: 956
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by bakerser View Post
Maybe you really don't mean to be throwing people off the track towards good retouching
Are you able to demonstrate on an image the nightmarish results one can expect from using his technique/tool relative to your own? That more than anything would save us all a lot of time/words/confusion.
Reply With Quote top
  #89  
Old 11-02-2010, 08:46 AM
bakerser bakerser is offline
Member
 
Join Date: Jan 2008
Posts: 91
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by Flashtones View Post
Are you able to demonstrate on an image the nightmarish results one can expect from using his technique/tool relative to your own? That more than anything would save us all a lot of time/words/confusion.
Are you familiar with the Highpass Sucks thread? If so, that's the problem with this approach. I can't post samples while mobile, but there are countless ones in that thread (if you've seen it).
Reply With Quote top
  #90  
Old 11-02-2010, 09:07 AM
Flashtones Flashtones is online now
Senior Member
 
Join Date: Apr 2009
Posts: 956
Re: Automatic Skin Smoothing.

Quote:
Originally Posted by bakerser View Post
Are you familiar with the Highpass Sucks thread? If so, that's the problem with this approach. I can't post samples while mobile, but there are countless ones in that thread (if you've seen it).

Familiar with the thread. Have read through it's entirety and played with most of the actions/scripts from there, and some very good ones developed here too (see work by Cain). I'm a big fan of your work and efforts to inform.

Not a math guy though, so I'm not one to count angels on the head of a pin - I'm in it for the visuals.

I understand the progress your technique yields over HP, particularly in making a base split that is visually equal to the original. What's less clear to me is if that advantage carries over as obviously in the Invert/Degrunge utilization, where contrast suppression is typically a desired effect.

I have yet to see side-by-side comparisons of say, your IHP action vs Shadowlight's tools. If it's advantages are obvious it shouldn't be hard to demonstrate visually. Doing so would go a long way here.
Reply With Quote top
Reply

  RetouchPRO > Technique > Photo Retouching


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is Off
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
Glamour retouch challenge here!!!! superkoax Photo Retouching 332 02-08-2011 09:52 PM
Automatic Skin Tone Retouching - Seeking feedback Pixarra Photo Retouching 15 11-09-2010 01:38 AM
Skin Smoothing Technique (Again) Novi Photo Retouching 10 10-16-2008 04:25 PM
need help doing pageant eyelashes and skin smoothing w/ photoshop cs sandylavallie Photo Retouching 3 05-06-2005 08:53 AM
skin smoothing action john_opitz Photo Retouching 10 08-21-2002 10:48 PM


All times are GMT -6. The time now is 01:04 PM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
SEO by vBSEO ©2011, Crawlability, Inc.
Copyright © 2016 Doug Nelson. All Rights Reserved