Table of Contents

This is the first part of a multi-part tutorial on blendshapes.

Some background on Blend shapes.

Well most animation programs out there come from a core that time is linear and you need to key certain values(usual XYZ coordinates) on the time line, to get an object to animate. well blendshape also work on the same fundamentals. the only difference being that they all have their own local(component) space to work with, which is the reason why they are so widely used for facial expressions, as they can work under the hierarchy of the main object or base, giving the ability to slow down, speedup, or hold a certain expression for a given duration of frames independent of the main object.

Each blendshape works from a value of 0 to 1, 0 being the shape of the base mesh or no-influence, and 1 being the target shapes or full-influence.
Blendshapes work in a linear format, that is it only holds information of the start position and the end position or in computer terms that is the start and the end vector, and the smallest route between those 2 positions is a straight line, which is why 2 key frame are always linear, unless given more advanced logic like ease in- and ease out etc. but sadly Maya`s blendshape has no such thing :( and therefore there is something that is called in-between targets, which is basically like putting another key frame in between the 0 to 1 value. This value can be anything in between 0.001 to 0.999, with all floats included.
This is usually the case when u need to make eyelids blinks for Bulgy Eyeballs or you need some similar kind of curvature attenuation in your blendshapes.

In Maya, the object that u want to deform is called the base object while the objects that you use to deform the shape of other objects are called target objects.
There can only be 1 base object for multiple target objects.

Creating Basic Objects for Blending.

For a basic trial, Take a sphere with subdiv axis and height of 8.This will be your base object.
Now duplicate it and move it to the left. Here you can modify the vertices of the object, to form a different shape.
Note: You cannot add loops or delete loops or edge now on this model. You have to work with the number that you currently have.
Note: You can use many deformers like clusters and lattice to make the desired target shapes.

Below is a base object, which is a sphere with 8×8 No of CVs and to the right are four target objects that i have modeled from the sphere.

• Create Blendshapes.

To create a new blendshape node,
Select all the target objects that you want to use as blends, and finally select the base object.

In the Animation Menu Set > Create Deformers > Blend Shape Option Box
and you should see a window open like below.

Create Blendshape Options. –

BlendShape Node – Give a name for the blendshape deformer node. (Default is blendShape).

Envelope - This is the nods scale factor of influence (Default is 1).

Orgin - Specifies if u want to take local space values of world space values from the targets.

Local Space will only respect component space value changes made to the target. All translate, rotations and scaling done to the object are ignored,widely used for making facial expression, and corrective shapes.
World Space will respect all changes from the target shapes including its translate, rotations and scaling, ive never found the need to use this. :)

Inbetween - If inbetween is checked on, all the selected targets will blend in series on the same slider, if it is checked off each of the targets will get their own individual sliders, so that  each target can be blended in parallel.

Above is the blenshape editor window, on the left is the view with inbetween turned on ie. a series style blend, and on the right is with inbetween turned off a parallel style blend.

Note : if u have multiple targets and inbetween is turned on, maya will place the inbetweens at possitions divided by the no of inbetweens. ie, if u have 5 inbetweens, and inbetween option is on, then the inbetweens will be placed at 5 positions from 0 to 1 at increments of 0.2

Check-Topology - If turned on it will check with the base and target topology is a match, if a match is not found it will not add the blendshape. Turn this option off if the vertex counts between the base and the targets vary.

Delete targets - If turned on this will delete the target shapes objects from the scene after it has been added onto the base as a belendshape. After this no further edits can be made to the blendshape targets therfore it is recommened to keep this option off.

After you are done with setting your desired options you can click on apply, and a new blenshape node will be created on the base object. To check the functionality of the blendshape u can check the blendshape editor by going into. Window > Animation Editors > Blendshape Editor you should see something similiar to the window that we saw in the above image for the parallel blender mode.

• Adding Blendshapes

So, if u have create some shapes and added them to the base object like shown above. But now you realize that you need more shapes, we can do the create blendshape thing again, since that creates a new node, and its best to keep a single blendshape node per base, so if u need to add more target shapes to the already existing blendshape node. we need to use the

Animation Menu Set > Edit Deformers > Blend Shape > Add OptionBox

Add Blendshape Options. -

Specify node -  Turn on, if u have multiple blendshapes on teh same selected mesh.

BlendShape node - This is activated, if the specify node option is checked. Its a checkbox to specify the blendshape Name to which u want to add the target shapes.

Existing Nodes – This is a drop down list of all available blenshapes in the scene. Selecting one will populate the blendshape node textfield above.

Click Apply and these are the only options that u need to set to add new blends to an already existing blendshape node.

• Adding Blendshapes as In-Between`s

Animation Menu Set > Edit Deformers > Blend Shape > Add OptionBox

Sometimes when making blendshapes, the blending does not happen as desired, or is too linear, or intersects the mesh at a certain point, in this case we can use inbetweens. Inbetweens are basically getting a third intermediate target shape which will blend in series with the base and the target shape at value 1. A inbetween can be placed anywhere in between 0.001 to 0.999, depending on the result required.

Add inbetween target – Turn on, to add the selected target as an inbetween shape.

Target Index – This is a little tricky, the target index is basically the number at which the blendshape is in the stack. If u do it manually u need to sit and count at what position is the blendshape thatyou want to add the inbetween to. or you can use this script that i made which adds numbering to all the blendshapes in a stack.
can be found here http://www.artbycrunk.com/blog/2009/03/scripts-crbseassist/ . Below is what it will look like with the script.

Inbetween weight - This is a value between 0.001 to 0.999. It specifies at what position on the slider the selected target should blend in.

Click Apply and scrub the slider in the blendshape editor, this should show you the inbetween blended target.

• Removing Blendshapes

Removing shapes from the blendshape node is nearly smiliar to adding shapes. you need to select the target shape that you want to delete, and then select the base object. and go to Animation Menu Set > Edit Deformers > Blend Shape > Remove OptionBox . Here u need to specify the blendshape node, and the target like explained above,and then click apply, the selected node will be removed.
This tutorial is quite tricky to explain with images and text, that’s why ill be making a screencast(video) in sometime, until then please give your inputs or query for the betterment of this tutorial.

Animation Menu Set > Edit Deformers > Blend Shape > Add OptionBox

People have continously kept asking me how Ive got the effect of rendering a wireframe over the ambient occlusion. and this is what this tutorial is all about.

• 1.  Add the Wireframe to all the required meshes by using the Rendering Wireframe Using ToonShader Tutorial or the Script.

• 2. Select all the polygon objects and make a new layer using the Create New Layer and Assign Selected Objects button in the Render Layers Section (see image below, the button with the blue sphere) . Name it AO_Layer.
• 3.  Select all the polygon objects and the Toon Wireframe objects and make another new layer. Name it Wireframe_Layer.

• 4. Select the AO_layer, If you are on maya 2008, u can right click and assign a preset on Ambient Occlusion, if your on 2009 then here is the Tutorial to make the shader and assign it to the the layer.
• 5.  Select the Wireframe_Layer.  Create a UseBackground Shader, Select all the polygonal objects in the layer and assign them the newly created UseBackground shader.

• 6.  Make sure in your Render Settings your renderer is set to MentalRay.

Currently one of the problem you will face is that the toonWireframe will only render in MayaSoftware renderer and the Ambient Occlusion will only render is Mental Ray. so to solve this issue we need to allocate which layer should use which renderer.

• 7. Select the Wireframe_Layer. Open the Render Settings and right click on render Using and Click on Create Layer Override. Then Slect Maya Software from the List.

• 8. Now, do a batch render, and both the render layers will render as separate image files, which u can then composite together.
• In case you do not have a composting package maya can do a composite of the layers for you. In the Render Layers Tab, Options-> Render All Layers and makes sure that in the render all layers options, the keep image mode option is set to composite layers.

• Here is what the final render looks like with the ambient occlusion render overlapped by the wireframe.

This model is something i had made back in 2005, its a computer from fallout2, i was just trying to make props for games. :)

Occlusion is a fast and simple method to simulate the effects of global illumination. Ambient occlusion is most often calculated by casting rays in every direction from the surface. Rays which reach the background or “sky” increase the brightness of the surface, whereas a ray which hits any other object contributes no illumination. As a result, points surrounded by a large amount of geometry are rendered dark, where as points with little geometry on the visible hemisphere appear light. The soft appearance achieved by ambient occlusion alone is similar to the way an object appears on an overcast day. So what basically this shader does is try to re-create the effect of ambient occlussion. It is moderately low on render time, and has significant advantanges in the final look of an image. The mib_amb_occlusion node come with a variety of attributes to customize the final rendered AO image.

First off the ambient occlussion node is a mental ray node so, so u would require the mayatomr plugin to be enabled. The ambient occlusion node can be found in the Hypershade under Create > MentalRay Textures > mib_amb_occlusion.

There are various ways to connect the mib_amb_occlusion node to the desired object.

1. The first is the preset that comes with maya`s render layer system, which makes a surface shader and connects the mib_amb_occlusion node to the surface shaders outColor.

2. In this case one can connect the mib_amb_occlusion node`s outValue to the ambient or Incandescence of any desired shader. The connetions are the same as above expect for the fact that u would not connect to outColor but to ambient or Incandescence.

3. One can also connect the mib_amb_occlusion node to the shading group of a material, the the shading groups attributes there will be a mental ray tab with a field called Material Shader.

__________________________________________________________________________________________

Samples

Controls the overall quality of the occlusion. Increasing the number of samples will reduce noise and give u a smoother render, the higher the number the more expensive the render cost. It quite easy to go overboard with this attribute, so know when to quit.

Default – 16

__________________________________________________________________________________________

Bright / Dark

These two sliders control the lightest/darkest color of the occlusion. It can work as a clamp to keep the black and white range under a limit or one can change the colors for other desired looks.

Default - White/Black

__________________________________________________________________________________________

Spread

Spread is the basic dropoff between the bright and the dark colors in the occlusion. A higher spread will require more samples.

Default - 0.800

__________________________________________________________________________________________

Max Distance

Max Distance is the how far a ray of occlusion will go in search of another object to occlude. Zero is basically all objects in infinity will occlude each other.

Default – 0.800

__________________________________________________________________________________________

Falloff

This attribute is only useable if the max distance is not eqaul to zero. It is basicall how fast the occulussion effect with die out with have over the distance.

Default – 1

__________________________________________________________________________________________

Reflective

When turned on, the occlussion will reflect on surronding objects. This is good when u have a lot of reflective objects in the scene.

Default – Off

__________________________________________________________________________________________

Output Mode

This constits of three modes, that defines what the returned color will be.

Default – 0

__________________________________________________________________________________________

Occlusion In Alpha

When turned on, the occlussion will show up in the alpha channel, For this to work u will have to use the 3rd step in connecting the shader to the ShadingGroup Node.
Turn on Pass Custom Alpha Channel  in the Render Settings > MentalRay Tab  > Under Custom Entities.

__________________________________________________________________________________________

Occlusion & Transparency

By default occlusion does not respect transparecy, then if u need transparent objects to be occluded, u will need to connect the AO shader to the amient or incandescence.

__________________________________________________________________________________________

miLabel Attribute

For the next two options to work  i.e id_inclexcl and id_nonself, all your objects need to be assigned with id numbers, and these id numbers need to be added in a new attribute called miLabel. So for the below to options to work you will need to add an attribute to each and every geometry in the scene, and give it an id number.

MEL SCRIPT

I have made a script for this purpose called crAssignID that lets u add the attribite and assingn id numbers to a group of selected objects.
It has a couple of helpfull options and can be found below, or looking into the scripts section of this blog.

http://www.artbycrunk.com/blog/2009/03/scripts-crassignid/

__________________________________________________________________________________________

id_inclexcl

This is basically an include/exclue from occlusion.
0 means that this attribute will have no effect. Same value as the miLabel means that only those objects can cause occlusion.
A Negative value of the miLabel means that those objects will not cause any occlusion.

id_nonself

This is basically an self or non self occlussion option again
0 means this atttribute will have no effect. Same value as the miLabel means that those object will not self occlude.

This is difficult to explain how these two attributes really works, you will just need to try it out yourself. If u have any problems please leave a comment :)

__________________________________________________________________________________________

Well, everyone knows the old method of rendering wireframes in maya through vector rendering, and rendering creased line, one set back to that was that it would tend to render each quad divided into 2 tris..  Well my major problem was with maya 2008 64bit doesnt suport vector rendering plugins. :) so i thought id try something new.. i mean there has to be another sollution out there.. hunted the net for sometime.. and came across this Highend Wireframe Tutorial, but in the modelling stage the model would not have UVs so this is not an efficient enough method, so after fidling around with the toon shader i gots it :)

Step 1.
Selct all ur objects and apply a toon shader outline.

Step 2.
In the PFXToonShape Turn ProfileLines – Off, Border Lines – Off and Crease Lines – On.

Step 3.
In the Crease Lines tab, Hard Creases Only – Off and Crease Min/Max Value to 0.

Step 4.
Adjust the Crease Line Width depending on how far or near ur model is to the camera.

Step 5.
Render Using Maya Software.

The renders are very low cost, and dont require any uv information, and if u smooth the mesh, the toon shader will automatically update itself with the new mesh topology. :)

And here is some stuff that i rendered in the same technique.
This is a truck from the SWAT KATS series that i had made around Oct-Nov 2007.

__________________________________________________________________________________________

MEL SCRIPT

I have made a script for this purpose called crToonWireframeRen that lets u add the wireframe and makes the settings that are required.
What u get is a render ready version of the wireframe on any selected meshes.

It is quite helpfull and can be found below, or looking into the scripts section of this blog.

http://www.artbycrunk.com/blog/2009/03/scripts-crtoonwireframeren/

__________________________________________________________________________________________

__________________________________________________________________________________________