Reading mode: Light Dark

Stair Interpolation resize methods compared..

Go to last post
This is an interesting problem. The most common means for upsizing are: one single up upsize in PS, stair interpolation, genuine fractals, and spline interpolation. Recently a better method has been added for those working with RAW files using ACR. Comparisons can be difficult, as there is no standardization of methodology. For instance, genuine fractals does not give good results if one starts with small files, but is great for changing a 40 MB file into a 120 MB file. Spline works best with smaller files, but gives annoying artifacts with large files. I use ACR myself, but if I were to do it in PS, I would use stair interpolation. Good luch with your project.
VG said:
A work in progress as I plan on adding much more info. As it is, I
welcome feedback, suggestions and/or questions about using such a
methodology.
http://www.vizualgroove.com/digitalimaging/sitest.htm
Regards,
VG
Click to expand...
 
VG,

What's your rationale for the predominantly irrational-numbered step percent values? ;-)
 
So you would ask....hehehe...just kiddin'
1/3 of 100% binaried down
VG
ZMR said:
VG,

What's your rationale for the predominantly irrational-numbered
step percent values? ;-)
Click to expand...
 
This isn't really a technical question as much as a curious one. I'm struck by the differences between the two when viewed several feet from the monitor. PS seems to have a longer tonal gradation resulting in a smoother appearance. Slide Click has more contrast which gives the appearance of extra sharpness.

Having never enlarged a digital image past 8x10, I'm wondering what the results would look like in a blown up image. Which looks best, the one with more contrast or the one with less?

--
Pam
http://www.pbase.com/pam_r
'art is working on something 'til you like it...then leaving it that way'
VG said:
A work in progress as I plan on adding much more info. As it is, I
welcome feedback, suggestions and/or questions about using such a
methodology.
http://www.vizualgroove.com/digitalimaging/sitest.htm
Regards,
VG
Click to expand...
 
ZMR said:
I was half serious, VG. I suppose there is an optimum value for the
number of steps given an enlargement ratio. How do you pick that
optimum step size?
Click to expand...
And would it make sense to have increasingly larger step sizes?
 
ZMR said:
I was half serious, VG. I suppose there is an optimum value for the
number of steps given an enlargement ratio. How do you pick that
optimum step size?
Click to expand...
If you're interested in a the theory behind stair interpolation you can start from http://citeseer.nj.nec.com/wolberg93fast.html . Just open the page and choose "Cached: PDF" from the top right part of the page.

It's a paper from George Wolberg. The conclusion is that with a well done stair interpolation you can save a lot of time and lose little quality respect to a correctly interpolated single step.

It requires some mathematical knowledge.

--
Fulvio Senore
 
Fulvio,

Thanx for the link. I though I had read about everything there was but this is good. I wish PS had a way to make a custom kernel for its bicubic interpolation routine. Something along the lines of the custom convolution kernel.
VG
Fulvio Senore said:
ZMR said:
I was half serious, VG. I suppose there is an optimum value for the
number of steps given an enlargement ratio. How do you pick that
optimum step size?
Click to expand...
If you're interested in a the theory behind stair interpolation you
can start from http://citeseer.nj.nec.com/wolberg93fast.html . Just
open the page and choose "Cached: PDF" from the top right part of
the page.

It's a paper from George Wolberg. The conclusion is that with a
well done stair interpolation you can save a lot of time and lose
little quality respect to a correctly interpolated single step.

It requires some mathematical knowledge.

--
Fulvio Senore
Click to expand...
 
At this point, I'm experimenting with even divisions that most people use in their actions. I plan on trying pretty much everything just about every way. So far, there aren't a lot of differences in most any of them past 3%
VG
ZMR said:
ZMR said:
I was half serious, VG. I suppose there is an optimum value for the
number of steps given an enlargement ratio. How do you pick that
optimum step size?
Click to expand...
And would it make sense to have increasingly larger step sizes?
Click to expand...
 
Pam,

You just won the 'I Ain't Be No Photoshop Mowron' award that is given out to people with common sense in digital imaging. I was really hoping that someone would finally get on this aspect.

You are indeed correct as to your visual curiosity. I call it 'perceived perception' in that an image can only be perceived to be ideal based on its ouput medium. Our monitors can't show what a continuous tone or even an inkjet printer can lay down on paper. Therefore, until the image is actually printed on a larger medium and viewed from an optimal distance, we can't exactly go along with WYSIWYG.

Side Click generated all 4 of the crops that you see, even the 10% one. Side Click uses Adobe Photoshop Scripting language which basically just tells PS what to do. It in itself doesn't do the actual interpolating.

As to which one would look the best, that is one of the future tests intended. I'm going to put crops from each percent method on a 20x30 and have it lab printed. I'll let ya'll know how it goes.
Good observation,
VG
Pam R said:
Having never enlarged a digital image past 8x10, I'm wondering what
the results would look like in a blown up image. Which looks best,
the one with more contrast or the one with less?

--
Pam
http://www.pbase.com/pam_r
'art is working on something 'til you like it...then leaving it
that way'
VG said:
A work in progress as I plan on adding much more info. As it is, I
welcome feedback, suggestions and/or questions about using such a
methodology.
http://www.vizualgroove.com/digitalimaging/sitest.htm
Regards,
VG
Click to expand...
Click to expand...
 
Fulvio Senore said:
ZMR said:
I was half serious, VG. I suppose there is an optimum value for the
number of steps given an enlargement ratio. How do you pick that
optimum step size?
Click to expand...
If you're interested in a the theory behind stair interpolation you
can start from http://citeseer.nj.nec.com/wolberg93fast.html . Just
open the page and choose "Cached: PDF" from the top right part of
the page.

It's a paper from George Wolberg. The conclusion is that with a
well done stair interpolation you can save a lot of time and lose
little quality respect to a correctly interpolated single step.

It requires some mathematical knowledge.

--
Fulvio Senore
Click to expand...
Thanks Fulvio, I will try to read it tonight, and I hope that I can understand it. VG's post scared me with some big words from that paper.
 
VG said:
At this point, I'm experimenting with even divisions that most
people use in their actions. I plan on trying pretty much
everything just about every way. So far, there aren't a lot of
differences in most any of them past 3%
VG
Click to expand...
Intuitively, too many steps would results in degradation too.
 
ZMR said:
Fulvio Senore said:
ZMR said:
I was half serious, VG. I suppose there is an optimum value for the
number of steps given an enlargement ratio. How do you pick that
optimum step size?
Click to expand...
If you're interested in a the theory behind stair interpolation you
can start from http://citeseer.nj.nec.com/wolberg93fast.html . Just
open the page and choose "Cached: PDF" from the top right part of
the page.

It's a paper from George Wolberg. The conclusion is that with a
well done stair interpolation you can save a lot of time and lose
little quality respect to a correctly interpolated single step.

It requires some mathematical knowledge.

--
Fulvio Senore
Click to expand...
Thanks Fulvio, I will try to read it tonight, and I hope that I can
understand it. VG's post scared me with some big words from that
paper.
Click to expand...
Heres a word that will make you run for the hills...Ronkonkoma
 
anyone have an action to do staircase.............?

It will save me the trouble of writing my own.
 
nuyah said:
anyone have an action to do staircase.............?

It will save me the trouble of writing my own.
Click to expand...
It is better to make your own. You need several, and they have to allow you to end up with a target size from a given source. Everyone’s needs are different and the actions are really easy to make.

I have settled on 5% increments. I start an action named 5 X 5 and resize an image 5 times at 5%. I start another action named 10 X 5 and run the first action 2 times. I make another called 20 X 5 and run the second action twice. Those three actions usually cover my normal resizing. I can run them in tandem or just combine them to make a new action if my needs change.

Just make sure to use the % in the resize so it will work on any image.

Someone suggested throwing in a little UM with SI. I might try a little after the first 5 so that I apply some UM each 5 steps.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top