A model for adding declarative rasterbased rendering effects to a 2D graphics environment is presented. As a result, the expressiveness of the traditional 2D rendering model is greatly enhanced, while still preserving the device independence, scalability, and high level geometric description of the underlying graphics.
On the Web, many graphics are presented as bitmap images in gif, jpg, or png format. Among the many disadvantages of this approach is the general difficulty of keeping the raster data in sync with the rest of the Web site. Many times, a web site designer must resort to a bitmap editor to simply change the title of a button. As the Web gets more dynamic, we desire a way to describe the "piece parts" of a site in a more flexible format. This chapter describes SVG's declarative filter effects model, which when combined with the 2D power of SVG can describe much of the common artwork on the web in such a way that clientside generation and alteration can be performed easily.
The filter effects model consists of a set of filtering operations (called "processing nodes" in the descriptions below) on one or more graphic primitives. Each processing node takes a set of graphics primitives as input and generates a new set of graphics primitives as output. After all processing nodes have completed, the resulting set of graphics primitives are returned and then rendered to the target media.
For example, a simple filter could replace one graphic by two  by adding a black copy of original offset to create a drop shadow. In effect, there are now two layers of graphics, both with the same original set of graphics primitives. In this example, the bottommost shadow layer could be blurred and become a raster layer, while the topmost layer could remain as higherorder graphics primitives (e.g., text or vector objects). Ultimately, the two layers are composited together and rendered into the background.
Filter effects introduce an additional step into the traditional 2D graphics pipeline. Consider the traditional 2D graphics pipeline:
Vector graphics primitives are specified abstractly and rendered onto the output device through a geometric transformation called the current transformation matrix, or CTM. The CTM allows the vector graphics code to be specified in a device independent coordinate system. At rendering time, the CTM accounts for any differences in resolution or orientation of the input vector description space and the device coordinate system. According to the "painter's model", areas on the device which are outside of the vector graphic shapes remain unchanged from their previous contents (in this case the droplet pattern).
Consider now, altering this pipeline slightly to allow intermediate filtering operations performed on the graphics primitives, producing a modified set of graphics primitives. The modified set of graphics primitives are rendered onto the output device in a second pass (??? This picture needs to be updated to show vector effects in addition to raster effects):
Many of the filter effects processing nodes operate on raster images only. If a given processing nodes requires an image as input, then all graphics elements are converted into an intermediate continuous tone raster image (a process called rasterization), which is then passed to the given processing node for processing. For those cases where rasterization is necessary, we introduce a new transformation matrix called the Effect Transformation Matrix, or ETM. Whenever it is necessary to perform raster effects, the graphics primitives will need to be rendered (i.e., rasterized) via the ETM onto the intermediate continuous tone image, using the ETM as the user space to image space transformation matrix. Ultimately, this intermediate continuous tone image will be rendered onto the output device using the standard 2D imaging path via a modified transform, CTM', such that the net effect of ETM followed by CTM' is equivalent to the original CTM. It is important to note that the filter effects intermediate continuous tone image contains coverage information so that non rendered parts of the original graphic are transparent in the intermediate image and remain unaffected on the output device, as required by the painter's model. A physical analog to this process is to imagine rendering the vector primitives onto a sheet of clear acetate and then transforming and overlaying the acetate sheet onto the output device. The resulting imaging model remains as deviceindependent as the original one, except we are now using the 2D imaging model itself to generate images to render.
So far, we really haven't added any new expressiveness to the imaging model. What we have done is reformulated the traditional 2D rendering model to allow an intermediate phase. However, now we can extend this further by allowing the application of vector and raster filter effects operations on the intermediate graphics primitives, still without sacrificing device independence. After applying the filter effects operations, the resulting set of graphics primitives are then rendered onto the device in the same way as above. (??? This picture needs to be updated to show vector effects in addition to raster effects).
In the picture above, the intermediate set of graphics primitives was processed in two ways. First a simple bump map lighting calculation was applied to add a 3D look, then another copy of the original layer was offset, blurred and colored black to form a shadow. The resulting transparent layers were then rendered via the painter's model onto the output device.
Filter effects are defined by a <filter> element with an associated ID. Filter effects are applied to elements which have a filter: property which reference a <filter> element. Here is an example:
<?xml version="1.0" standalone="no"?> <!DOCTYPE svg PUBLIC "//W3C//DTD SVG June 1999//EN" "http://www.w3.org/Graphics/SVG/svg19990625.dtd"> <svg width="4in" height="3in"> <defs> <filter id="CoolTextEffect"> <! Definition of filter goes here > </filter> </defs> <text style="filter:url(#CoolTextEffect)">Text with a cool effect</text> </svg>
When applied to grouping elements such as <g>, the filter: property applies to the contents of the group as a whole. The effect should be as if the group's children did not render to the screen directly but instead just added their resulting graphics primitives to the group's graphics display list (GDL), which is then passed to the filter for processing. After the group filter is processed, then the result of the filter should be rendered to the target device (or passed onto a parent grouping element for further processing in cases such as when the parent has its own group filter).
The <filter> element consists of a sequence of processing nodes which take a set of graphics primitives as input, apply filter effects operations on the graphics primitives, and produce a modified set of graphics primitives as output. The processing nodes are executed in sequential order. Here is an example:
<?xml version="1.0" standalone="no"?> <!DOCTYPE svg PUBLIC "//W3C//DTD SVG June 1999//EN" "http://www.w3.org/Graphics/SVG/svg19990625.dtd"> <filter id="Shadow"> <REGaussianBlur in="SourceAlpha" radius="3" nodeid="blurredAlpha" /> <REOffset in="blurredAlpha" dx="2" dy="1" nodeid="offsetBlurredAlpha" /> <REDiffuseLighting in="blurredAlpha" diffuseconstant=".5" nodeid="bumpMapDiffuse" /> <REComposite in="bumpMapDiffuse" in2="SourcePaint" operator="arithmetic" k1="1" nodeid="litPaint" /> <RESpecularLighting in="blurredAlpha" specularconstant=".5" specularexponent="10" lightcolor="distantlight" nodeid="bumpMapSpecular" /> <REComposite in="litPaint" in2="bumpMapSpecular" operator="arithmetic" k2="1" k3="1" nodeid="litPaint" /> <REComposite in="litPaint" in2="SourceAlpha" mode="AinB" nodeid="litPaint" /> <REMerge> <REMergeNode in="litPaint" /> <REMergeNode in="offsetBlurredAlpha" /> </REMerge> </filter> <text style="fontsize:36; fill:red; filter:url(#Shadow)" x="10" y="250">Shadowed Text</text> </svg>
For most processing nodes, the in (and sometimes in2) attribute identifies the graphics which serve as input and the nodeid attribute gives a name for the resulting output. The in and in2 attributes can point to either:
The default value for in is the output generated from the previous processing node. In those cases where the output from a given processing node is used as input only by the next processing node, it is not necessary to specify the nodeid on the previous processing node or the in attribute on the next processing node. In the example above, there are a few cases (show highlighted) where nodeid and in did not have to be provided.
Filters do not use XML IDs for nodeids; instead, nodeids can be any arbitrary attribute string value. nodeids only are meaningful within a given <filter> element and thus have only local scope. If a nodeid appears multiple times within a given <filter> element, then a reference to that nodeid will use the closest preceding processing node with the given nodeid. Forward references to nodeids are invalid.
A <filter> element can define a region on the canvas on which a given filter effect applies and can provide a resolution for any intermediate continuous tone images used in to process any rasterbased processing nodes. The <filter> element has the following attributes:
xpixels [ypixels])
and indicates
the width/height of the intermediate images in pixels. If not provided,
then a reasonable default resolution appropriate for the target device
will be used. (For displays, an appropriate display resolution,
preferably the current display's pixel resolution, should be the default.
For printing, an appropriate
common printer resolution, such as 400dpi, should be the default.)For performance reasons on display devices, it is recommended that the filter effect region is designed to match pixelforpixel with the background pixmap.
It is often necessary to provide padding space because the filter effect might impact bits slightly outside the tightfitting bounding box on a given object. For these purposes, it is possible to provide negative percentage values for x, y and percentages values greater than 100% for width, height. For example, x="10%" y="10%" width="110%" height="110%".
It is possible to define a generalized template filter and then invoke it multiple times with slightly different parameters using the filter effect's parameter substitution features. Filter parameters are defined by defining a given attribute value within a processing node to be:
To illustrate:
<?xml version="1.0" standalone="no"?> <!DOCTYPE svg PUBLIC "//W3C//DTD SVG June 1999//EN" "http://www.w3.org/Graphics/SVG/svg19990625.dtd"> <svg width="4in" height="3in"> <defs> <filter id="UseParams"> <REImageTurbulence basefrequency="{$freq}" numoctaves="{$num}" /> </filter> </defs> <rect x="70" y="10" width="40" height="40" style="fill:red; filter:url(#UseParams); filterparams:freq(20) num(10)" /> </svg>
The following two attributes are available for all processing nodes (the exception is REMerge and REImage, which do not have an in attribute):
Common Attributes 
nodeid Assigned name for this node. If supplied, then the GDL resulting after processing the given node is saved away and can be referenced as input to a subsequent processing node. in If supplied, then indicates that this processing node should use either the output of previous node as input or use some standard keyword to specify alternate input. Available keywords include:

The following is a catalog of the individual processing nodes. Unless otherwise stated, all image filters operate on linear premultiplied RGBA samples. Filters which work more naturally on non premultiplied data (ColorMatrix and ComponentTransfer) will temporarily undo and redo premultiplication as specified. All raster effect filtering operations take 1 to N input RGBA images, additional attributes as parameters, and produce a single output RGBA image.
NodeType 
ColorMatrix 
Processing NodeSpecific Attributes 
type, string (one of: matrix, saturate, huerotate, luminancetoalpha) values

Description 
This filter performs  R'   a00 a01 a02 a03 a04   R   G'   a10 a11 a12 a13 a14   G   B'  =  a20 a21 a22 a23 a24  *  B   A'   a30 a31 a32 a33 a34   A   1   0 0 0 0 1   1  for every pixel. The RGBA and R'G'B'A' values are automatically nonpremultiplied temporarily for this operation. The following shortcut definitions are provide for compactness. The following tables show the mapping from the shorthand form to the corresponding longhand (i.e., matrix with 20 values) form: saturate value (0..100) s = value/100  R'  0.213+0.787s 0.7150.715s 0.0720.072s 0 0   R   G'  0.2130.213s 0.715+0.285s 0.0720.072s 0 0   G   B'  = 0.2130.213s 0.7150.715s 0.072+0.928s 0 0  *  B   A'   0 0 0 1 0   A   1   0 0 0 0 1   1  huerotate value (0..360)  R'   a00 a01 a02 0 0   R   G'   a10 a11 a12 0 0   G   B'  =  a20 a21 a22 0 0  *  B   A'   0 0 0 1 0   A   1   0 0 0 0 1   1  where the terms a00, a01, etc. are calculated as follows:  a11 a12 a13  [+0.213 +0.715 +0.072]  a21 a22 a13  = [+0.213 +0.715 +0.072] +  a11 a12 a13  [+0.213 +0.715 +0.072] [+0.787 0.715 0.072] cos(huerotate value) * [0.212 +0.285 0.072] + [0.213 0.715 +0.928] [0.213 0.715+0.928] sin(huerotate value) * [+0.143 +0.1400.283] [0.787 +0.715+0.072] Thus, the upper left term of the hue matrix turns out to be: .213 + cos(huerotate value)*.787  sin(huerotate value)*.213 luminancetoalpha  R'   0 0 0 0 0   R   G'   0 0 0 0 0   G   B'  =  0 0 0 0 0  *  B   A'   0.299 0.587 0.114 0 0   A   1   0 0 0 0 1   1  
Implementation issues 
These matrices often perform an identity mapping in the alpha channel. If that is the case, an implementation can avoid the costly undoing & redoing of the premultiplication for all pixels with A = 1. 
NodeType 
ComponentTransfer 
Processing NodeSpecific Attributes  None. 
Processing NodeSpecific SubElements 

Description 
This filter performs componentwise remapping of data as follows: R' = Fr( R ) G' = Fr( G ) B' = Fr( B ) A' = Fr( A ) for every pixel. The RGBA and R'G'B'A' values are automatically nonpremultiplied temporarily for this operation. When type="table", the transfer function consists of a linearly interpolated lookup table. k/N <= C < (k+1)/N => C' = vk + (C  k/N)*N * (vk+1  vk) When type="linear", the transfer function consists of a linear function describes by the following equation: C' = slope*C + offset When type="gamma", the transfer function consists of the following equation: C' = amplitude*pow(C, exponent) + offset 
Comments 
This filter allows operations like brightness adjustment, contrast adjustment, color balance or thresholding. We might want to consider some predefined transfer functions such as identity, gamma, sRGB transfer, sinewave, etc. 
Implementation issues 
Similar to the ColorMatrix filter, the undoing and redoing of the premultiplication can be avoided if Fa is the identity transform and A = 1. 
NodeType 
REBlend  
Processing NodeSpecific Attributes 
mode, One of the image blending modes (see table below). Default is: normal in2, The second image ("B" in the formulas) for the compositing operation.  
Description 
This filter composites two objects together using commonly used highend imaging software blending modes. Performs the combination of the two input images pixelwise in image space.  
Implementation Notes 
The compositing formula, expressed using premultiplied colors: qr = qb*(1qa) + qa*(1qb) + qa*qb cr = (1qa)*cb + (1qb)*ca + qa*qb*(Blend(ca/qa,cb/qb)) where: qr = Result opacity cr = Result color (RGB)  premultiplied qa = Opacity value at a given pixel for image A qb = Opacity value at a given pixel for image B ca = Color (RGB) at a given pixel for image A  premultiplied cb = Color (RGB) at a given pixel for image B  premultiplied Blend = Image compositing function, depending on the compositing mode The following table provides the list of available image blending modes:

NodeType 
REColor 
Processing NodeSpecific Attributes 
color, RGBA color (floating point?) 
Description 
Creates an image with infinite extent filled with color 
NodeType 
REComposite 
Processing NodeSpecific Attributes 
operator, one of (over, in, out, atop, xor, arithmetic). Default is: over. arithmeticconstants, k1,k2,k3,k4 in2, The second image ("B" in the formulas) for the compositing operation. 
Description 
This filter performs the combination of the two input images pixelwise in image space. over, in, atop, out, xor use the PorterDuff compositing operations. For these operations, the extent of the resulting image can be affected. In other words, even if two images do not overlap in image space, the extent for over will essentially include the union of the extents of the two input images. arithmetic evaluates k1*i1*i2 + k2*i1 + k3*i2 + k4, using componentwise arithmetic with the result clamped between [0..1]. 
Comments 
arithmetic are useful for combining the output from the DiffuseLighting and SpecularLighting filters with texture data. arithmetic is also useful for implementing dissolve. 
NodeType 
REDiffuseLighting 
Processing NodeSpecific Attributes 
resultscale (Multiplicative scale for the result.
This allows the result of the DiffuseLighting nodeto represent
values greater than 1) 
Processing NodeSpecific SubElements 

Description 
Light an image using the alpha channel as a bump map. The resulting image is an RGBA opaque image based on the light color with alpha = 1.0 everywhere. The lighting caculation follows the standard diffuse component of the Phong lighting model. The resulting image depends on the light color, light position and surface geometry of the input bump map. Color or texture is mean to be applied via a multiply (mul) composite operation. Dr = (kd * N.L * Lr) / resultscale Dg = (kd * N.L * Lg) / resultscale Db = (kd * N.L * Lb) / resultscale Da = 1.0 / resultscale where kd = diffuse lighting constant N is a function of x and y and depends on the surface gradient as follows: The surface described by the input alpha image Ain (x,y) is: Z (x,y) = surfacescale * Ain (x,y) Surface normal is calculated using the Sobel gradient 3x3 filter: Nx (x,y)=  surfacescale * 1/4*(( I(x+1,y1) + 2*I(x+1,y) + I(x+1,y+1))  (I(x1,y1) + 2*I(x1,y) + I(x1,y+1))) Ny (x,y)=  surfacescale * 1/4*(( I(x1,y+1) + 2*I(x,y+1) + I(x+1,y+1))  (I(x1,y1) + 2*I(x,y1) + I(x+1,y1))) Nz (x,y) = 1.0 N = (Nx, Ny, Nz) / Norm((Nx,Ny,Nz)) L, the unit vector from the image sample to the light is calculated as follows: For Infinite light sources it is constant: Lx = cos(azimuth)*cos(elevation) Ly = sin(azimuth)*cos(elevation) Lz = sin(elevation) For Point and spot lights it is a function of position: Lx = Lightx  x Ly = Lighty  y Lz = Lightz  Z(x,y) L = (Lx, Ly, Lz) / Norm(Lx, Ly, Lz) where Lightx, Lighty, and Lightz are the input light position. Lr,Lg,Lb, the light color vector is a function of position in the spot light case only: Lr = Lightr*pow((L.S),specularexponent) Lg = Lightg*pow((L.S),specularexponent) Lb = Lightb*pow((L.S),specularexponent) where S is the unit vector pointing from the light to the point (pointsatx, pointsaty, pointsatz) in the xy plane: Sx = pointsatx  Lightx Sy = pointsaty  Lighty Sz = pointsatz  Lightz S = (Sx, Sy, Sz) / Norm(Sx, Sy, Sz) If L.S is positive no light is present. (Lr = Lg = Lb = 0) 
Comments 
This filter produces a light map, which can be combined with a texture image using the multiply term of the arithmetic <Composite> compositing method. Multiple light sources can be simulated by adding several of these light maps together before applying it to the texture image. 
NodeType 
REDisplacementMap 
Processing NodeSpecific Attributes 
scale 
Description 
Uses Input2 to spatially displace Input1, (similar to the Photoshop displacement filter). This is the transformation to be performed: P'(x,y) < P( x + scale * ((XC(x,y)  .5), y + scale * (YC(x,y)  .5)) where P(x,y) is the source image, Input1, and P'(x,y) is the destination. XC(x,y) and YC(x,y) are the component values of the designated by the xchannelselector and ychannelselector. For example, to use the R component of Image2 to control displacement in x and the G component of Image2 to control displacement in y, set xchannelselector to "R" and ychannelselector to "G". 
Comments 
The displacement map defines the inverse of the mapping performed. 
Implementation issues 
This filter can have arbitrary nonlocalized effect on the input which might require substantial buffering in the processing pipeline. However with this formulation, any intermediate buffering needs can be determined by scale which represents the maximum displacement in either x or y. 
NodeType 
REGaussianBlur 
Processing NodeSpecific Attributes 
stddeviation. 
Description 
Perform gaussian blur on the input image. The Gaussian blur kernel is an appoximation of the normalized convolution: H(x) = exp(x2/ (2s2)) / sqrt(2* pi*s2) where 's' is the standard deviation specified by stddeviation. This can be implemented as a separable convolution. For larger values of 's' (s >= 2.0), an approximation can be used: Three successive boxblurs build a piecewise quadratic convolution kernel, which approximates the gaussian kernel to within roughly 3%. let d = floor(s * 3*sqrt(2*pi)/4 + 0.5) ... if d is odd, use three boxblurs of size 'd', centered on the output pixel. ... if d is even, two boxblurs of size 'd' (the first one centered one pixel to the left, the second one centered one pixel to the right of the output pixel one box blur of size 'd+1' centered on the output pixel. 
Implementation Issues 
Frequently this operation will take place on alphaonly images, such as that produced by the builtin input, SourceAlpha. The implementation may notice this and optimize the single channel case. If the input has infinite extent and is constant, this operation has no effect. If the input has infinite extent and is a tile, the filter is evaluated with periodic boundary conditions. 
NodeType 
REImage 
Processing NodeSpecific Attributes 
href, reference to external image data. 
Description 
Refers to an external image which is loaded or rendered into an RGBA raster. If imagingmatrix is not specified, the image takes on its natural width and height and is positioned at 0,0 in image space. The imageref could refer to an external image, or just be a reference to another piece of SVG. This node produces an image similar to the builtin image source SourceGraphic except from an external source. 
NodeType 
REMerge 
Processing NodeSpecific Attributes 
none 
Processing NodeSpecific SubElements  Each <REMerge> element can have any number of <REMergeNode> subelements, each of which has an in attribute. 
Description 
Composites input image layers on top of each other using the over operator with Input1 on the bottom and the last specified input, InputN, on top. 
Comments 
Many effects produce a number of intermediate layers in order to create the final output image. This filter allows us to collapse those into a single image. Although this could be done by using n1 Compositefilters, it is more convenient to have this common operation available in this form, and offers the implementation some additional flexibility (see below). 
Implementation issues 
The canonical implementation of REMerge is to render the entire effect into one RGBA layer, and then render the resulting layer on the output device. In certain cases (in particular if the output device itself is a continuous tone device), and since merging is associative, it may be a sufficient approximation to evaluate the effect one layer at a time and render each layer individually onto the output device bottom to top. 
NodeType 
REMorphology 
Processing NodeSpecific Attributes 
operator,one of erode or dilate. 
Description 
This filter is intended to have a similar effect as the min/max filter in Photoshop and the width layer attribute in ImageStyler. It is useful for "fattening" or "thinning" an alpha channel, The dilation (or erosion) kernel is a square of side 2*radius + 1. 
Implementation issues 
Frequently this operation will take place on alphaonly images, such as that produced by the builtin input, SourceAlpha. In that case, the implementation might want to optimize the single channel case. If the input has infinite extent and is constant, this operation has no effect. If the input has infinite extent and is a tile, the filter is evaluated with periodic boundary conditions. 
NodeType 
REOffset 
Processing NodeSpecific Attributes 
dx,dy 
Description 
Offsets an image relative to its current position in the image space by the specified vector. 
Comments 
This is important for effects like drop shadow etc. 
NodeType 
RESpecularLighting 
Processing NodeSpecific Attributes 
surfacescale height of surface when Ain = 1. specularconstant ks in Phong lighting model. Range 0.0 to 1.0 specularexponent exponent for specular term, larger is more "shiny". Range 1.0 to 128.0. lightcolor RGB

Description 
Light an image using the alpha channel as a bump map. The resulting image is an RGBA image based on the light color. The lighting caculation follows the standard specular component of the Phong lighting model. The resulting image depends on the light color, light position and surface geometry of the input bump map. The result of the lighting calculation is added. We assume that the viewer is at infinity the z direction (i.e the unit vector in the eye direction is (0,0,1) everywhere. Sr = ks * pow(N.H, specularexponent) * Lr Sg = ks * pow(N.H, specularexponent) * Lg Sb = ks * pow(N.H, specularexponent) * Lb Sa = max(Sr, Sg, Sb) where ks = specular lighting constant N = surface normal unit vector, a function of x and y H = "halfway" unit vectorbetween eye unit vector and light unit vector Lr,Lg,Lb = RGB components of light See DiffuseLighting for definition of N and (Lr, Lg, Lb). The definition of H reflects our assumption of the constant eye vector E = (0,0,1): H = (L + E) / Norm(L+E) where L is the light unit vector. Unlike the DiffuseLighting, the SpecularLighting filter produces a nonopaque image. This is due to the fact that specular result (Sr,Sg,Sb,Sa) is meant to be added to the textured image. The alpha channel of the result is the max of the color components, so that where the specular light is zero, no additional coverage is added to the image and a fully white highlight will add opacity. 
Comments 
This filter produces an image which contains the specular reflection part of the lighting calculation. Such a map is intended to be combined with a texture using the add term of the arithmetic Composite method. Multiple light sources can be simulated by adding several of these light maps before applying it to the texture image. 
Implementation issues 
The DiffuseLighting and SpecularLighting filters will often be applied together. An implementation may detect this and calculate both maps in one pass, instead of two. 
NodeType 
RETile 
Processing NodeSpecific Attributes 
none 
Description 
Creates an image with infinite extent by replicating source image in image space. 
NodeType 
RETurbulence 
Processing NodeSpecific Attributes 
basefrequency 
Description 
Adds noise to an image using the Perlin turbulencefunction. It is possible to create bandwidthlimited noise by synthesizing only one octave. For a detailed description the of the Perlin turbulencefunction, see "Texturing and Modeling", Ebert et al, AP Professional, 1994. If the input image is infinite in extent, as is the case with a constant color or a tile, the resulting image will have maximal size in image space. 
Comments 
This filter allows the synthesis of artificial textures like clouds or marble. 
Implementation issues 
It might be useful to provide an actual implementation for the turbulence function, so that consistent results are achievable. 
NodeType 
VEAdjustGraphics 
Processing NodeSpecific Attributes 
transform a supplemental transformation (e.g., a scale, a rotate or a translate) on the current filter effects coordinate system style override CSS properties that applies to all objects in the current GDL 
Description 
Add a supplemental transformation (e.g., a scale, a rotate or a translate) on the current coordinate system or supply override CSS properties (e.g., a different fill or stroke property) that will apply to all objects in the current GDL 
NodeType 
VEPathTurbulence 
Processing NodeSpecific Attributes 
basefrequency 
Description 
Adds noise to a vector graphics using 1dimensional Perlin turbulencefunction. 
Implementation issues 
It might be useful to provide an actual implementation for the turbulence function, so that consistent results are achievable. 