Frei0r plugins Re: [gephex-user] Another new module: Infinite Impulse Response image filter

[binary millenium]

I built the existing frei0r plugins as dlls to use with the windows build of
gephex and put them here:
http://bmillenium.sourceforge.net/frei0r-plugins-dlls-1.1.19.zip

The generator and filter effects work, but the mixer2 effects all give me a
'invalid plugin_type' error when trying to use them with 0.4.3.  Are those
kinds of effects not supported in that version of gephex yet or did I build
them wrong?

-bm

On 1/15/07, JS <lord_tatien at yahoo.ca> wrote:
>
> Hi,
>
> It'd be nice if you could port those plugins into the Frei0r plugin
> framework. The framework was made mostly by the Gephex crew but is also
> supported by many other open source softwares. See this site for details:
> http://www.piksel.org/Frei0r
>
> Cheers,
>
> ++ js
>
> binary millenium wrote:
> > Here's another filter that implements an IIR filter on input video:
> >
> > http://bmillenium.sourceforge.net/zimagefiltermodule.zip
> >
> > Source and windows dlls provided, I'll provide linux x86 libs for this
> > and the earlier modules soon.
> >
> > The filter has two buffers of size four, one stores the current input
> > image and also buffers of the input image for three samples into the
> > past and also stores three copies of the output image for three
> > samples into the past.  8 coefficients sum all these buffered images
> > up into one combined new output frame (which is then buffered for use
> > in the next time step).
> >
> > The general equation where y is the output image, x is the input
> > image, and xn is the current image and xn-1 the image from one
> > timestep back is:
> > yn = (a0*xn + a1*xn-1 + a2*xn-2 + a3*xn-3  - b1*yn-1 - b2*yn-2 -
> > b3*yn-3) / b0
> >
> > If you know a thing or two about filter design you can make custom
> > low-pass, high-pass, or band-pass filters.
> >
> > It would be possible to extend the length of the buffers, but there's
> > a large performance hit just for the current size.  The code is
> > currently fairly ugly and I probably didn't make it as efficient as it
> > could be (it unnecessarily buffers an extra output image that is
> > unused for instance).
> >
> > -bm
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.gephex.org/pipermail/gephex-user/attachments/20070128/ffd2c5d6/attachment.htm