This is Interesting: Free Magazines for Graphics designers and webmasters  


Home > Archive > Computer Graphics with Photoshop > May 2006 > Converting acv file to a lut





You are viewing an archived Text-only version of the thread. To view this thread in it's original format and/or if you want to reply to this thread please [click here]

Author Converting acv file to a lut
Des

2006-05-07, 6:16 pm

Hello

Does anyone know of an algorithm which will accurately convert an acv file
to a look-up-table?

Reading an acv file is no problem - it's just a sequence of 2D points for
each curve.
But I'm having big problems converting the points to the same curve as
PhotoShop. I can get close - but not spot on.

Many thanks for any input or pointers.

Des


Mike Russell

2006-05-07, 6:16 pm

"Des" <DesG_nospam_@btclick.com> wrote in message
news:0L2dnUNxmfAI08PZnZ2dnUVZ8t2dnZ2d@bt.com...
> Hello
>
> Does anyone know of an algorithm which will accurately convert an acv file
> to a look-up-table?
>
> Reading an acv file is no problem - it's just a sequence of 2D points for
> each curve.
> But I'm having big problems converting the points to the same curve as
> PhotoShop. I can get close - but not spot on.
>
> Many thanks for any input or pointers.


Curve a gradient and save it as a raw file. Create an action to do this if
you have multiple curves to convert.

Splines should do a good job of duplicating the curve mathematically. What
function are you using, and what is the nature of the errors you are
encountering?
--
Mike Russell
www.curvemeister.com


Des

2006-05-07, 10:16 pm


"Mike Russell" <RE-MOVEmike@Curvemeister.comRE-MOVE> wrote in message
news:qws7g.69087$F_3.43191@newssvr29.news.prodigy.net...
> "Des" <DesG_nospam_@btclick.com> wrote in message
> news:0L2dnUNxmfAI08PZnZ2dnUVZ8t2dnZ2d@bt.com...
file[color=darkred]
for[color=darkred]
>
> Curve a gradient and save it as a raw file. Create an action to do this

if
> you have multiple curves to convert.
>
> Splines should do a good job of duplicating the curve mathematically.

What
> function are you using, and what is the nature of the errors you are
> encountering?
> --


Thanks for the response.

The PS curve is defined entirely by the input points (ie the points written
to or read from the acv file).
In addition the curve is not changed if you add another point that is
already on the curve - which is intuitively expected.

Taken together this means that the curve must maximise/minimise some global
property.
I've initially assumed that the curve is a minimum curvature curve.

I tried fitting a cubic bezier spline to the points with the control points
set to minimise curvature (which is equivalent to ensuring curvature
continuity for a cubic spline - ie a b-spline).

My results look good in terms of curvature, but are different from the curve
PS generates. The PS curve looks too loose to me. I was wondering if anyone
knew what sort of curve is being fitted - maybe they use a higher order
spline?

Thanks again
Des


Mike Russell

2006-05-08, 3:16 am

"Des" <DesG_nospam_@btclick.com> wrote in message
news:296dndaqX4kn7MPZnZ2dnUVZ8qKdnZ2d@bt.com...

[re spline curve for photoshop curves]

> The PS curve is defined entirely by the input points (ie the points
> written
> to or read from the acv file).
> In addition the curve is not changed if you add another point that is
> already on the curve - which is intuitively expected.
>
> Taken together this means that the curve must maximise/minimise some
> global
> property.
> I've initially assumed that the curve is a minimum curvature curve.
>
> I tried fitting a cubic bezier spline to the points with the control
> points
> set to minimise curvature (which is equivalent to ensuring curvature
> continuity for a cubic spline - ie a b-spline).
>
> My results look good in terms of curvature, but are different from the
> curve
> PS generates. The PS curve looks too loose to me. I was wondering if
> anyone
> knew what sort of curve is being fitted - maybe they use a higher order
> spline?


The cubic spline algorithm from the Numeric Recipes book is an exact match.
--
Mike Russell
www.curvemeister.com


Des

2006-05-08, 6:16 pm


"Mike Russell" <RE-MOVEmike@Curvemeister.comRE-MOVE> wrote in message
news:99y7g.12273$Lm5.8960@newssvr12.news.prodigy.com...
> "Des" <DesG_nospam_@btclick.com> wrote in message
> news:296dndaqX4kn7MPZnZ2dnUVZ8qKdnZ2d@bt.com...
>
> [re spline curve for photoshop curves]
>
>
> The cubic spline algorithm from the Numeric Recipes book is an exact

match.
> --


Thanks a million.

Yes, I implemented that and as you say it matches PS exactly.


Thanks again
Des


Waldo

2006-05-08, 6:16 pm

Des wrote:
> Hello
>
> Does anyone know of an algorithm which will accurately convert an acv file
> to a look-up-table?
>
> Reading an acv file is no problem - it's just a sequence of 2D points for
> each curve.
> But I'm having big problems converting the points to the same curve as
> PhotoShop. I can get close - but not spot on.
>
> Many thanks for any input or pointers.
>
> Des
>
>


Des,

How are the 2D points stored in the ACV file? Is the data type floats or
ints?

Waldo
Des

2006-05-08, 6:16 pm


"Waldo" <nobody@onbody.co> wrote in message
news:1147093481.259532@news-ext.oce.nl...
> Des,
>
> How are the 2D points stored in the ACV file? Is the data type floats or
> ints?
>
> Waldo



Curves file format

Length Description

2 Version ( = 1)
2 Count of curves in the file.

The following is the data for each curve specified by count above

2 Count of points in the curve (short integer from 2...19)

point count
4
Curve points.Each curve point is a pair of short integers where the
first number is the output value (vertical coordinate on the Curves
dialog graph) and the second is the input value. All coordinates have
range 0 to 255. See also "Null curves" below.

Hth
Des


Waldo

2006-05-08, 6:16 pm

Des wrote:
> "Waldo" <nobody@onbody.co> wrote in message
> news:1147093481.259532@news-ext.oce.nl...
>
>
>
>
> Curves file format
>
> Length Description
>
> 2 Version ( = 1)
> 2 Count of curves in the file.
>
> The following is the data for each curve specified by count above
>
> 2 Count of points in the curve (short integer from 2...19)
>
> point count
> 4
> Curve points.Each curve point is a pair of short integers where the
> first number is the output value (vertical coordinate on the Curves
> dialog graph) and the second is the input value. All coordinates have
> range 0 to 255. See also "Null curves" below.
>
> Hth
> Des
>
>


Thanks!

Waldo
Sponsored Links


Copyright 2003 - 2008 forum4designers.com  Software forum  Computer Hardware reviews