blathering bloggerskite

“Revise your model as needed to create an improved animation (as a .mov file) of your face acting out your dialog. This should be very close to the final version, with final lighting, surfacing and camera motion as well as refined expression and speech animation. “

Final is listed below. There’s still something wrong with the eyes…namely, I have somehow turned off the shinyness and still can’t find where to fix it. Otherwise, I’m pretty happy with the result.

Click here for .mov

“Revise your model as needed to create an improved animation (as a .mov file) of your face acting out your dialog. In this version begin to focus on lighting and surfacing as well as expression animation.”

A prettyish picture from the eventual render…with unintentional 5 o’clock shadow X( :

And a next-to finished version of the animation, sans surfacing.
Click here for .mov

“Using the refined, rigged version of your model create the initial animation (as a .mov file) of your face acting out your dialog.”

Finally….getting blendshapes to do what I want them to do, and putting some decent skin texture on there. Obviously this is not close to the final version, particularly since the hair needs shading of some sort, and none of the sound is synched. This is what happens when you realise your computer does not have working sound, so you make your rough animation based on the shape of the waveform….

Click here for .mov

Here are some of the first passes of blend shapes (made by rigging a face with clusters, pulling these, and refining the result before making it into a target shape):

…And some more expressions and visemes (slightly terrifying, here)…

What I do have at the moment is continued issues with combining blendshapes and joint rotations…which work okay separately, but together make parts of the model explode. Argh.

animation of neck twist

some of the blend shapes

The eyes, and especially the nose and mouth are more refined this time around…nostrils have magically appeared and the geometry around the lips is a vast improvement. Ears are still a figment of wildest fantasy and the hair, at this point, looks more like a turban…so there’s plenty still to do.

Refined nose and mouth:

Helmet-head in progress (+ earcup):

For the final facial animation, I am thinking of using dialogue something along these lines:

(In answer to ‘What happened to your body?’): Oh….well, I was a magician’s assistant.  Y’know that trick where they saw a woman in half?  …Er.    He missed.

Reference drawings:

Images of early model: The mouth especially needs to be redone, as the shape is good, but the geometry isn’t quite right for ideal speech.  The nose looks okay in profile/quarter view, but in full face seems overly wide or flat…which is probably a result of not having it right in the reference itself.  There are also a few odd polygons around the eyes and sides of face that need fixing, but some will be altered/hidden by the addition of ears and hair and may not matter too much.

…namely, showcasing the 14 lighting set-ups learned from portrait photography. (13 single-source, one with multiple lights)

Full-Face: Viewpoint directly in front of face

Lit hight above viewpoint & Lit above viewpoint

Lit at viewpoint & Lit below viewpoint

Lit at 45 degrees to the side and above & Lit directly at side of face (90 degrees)

Three-quarter pose: Face turned slightly away from viewpoint

Short lighting & Butterfly lighting & Broad lighting

Facial profiles: Viewpoint positioned directly to side of head

Lit directly at front of face & Lit at or close to viewpoint

Lit at angle from back & Lit in silhouette

Multiple-source lighting: Viewpoint at three-quarter pose, includes direct full-front of face lighting, side lighting, silhouette lighting, and slight backlighting (hairlight)

Usual four-light set-up

< Click here for the animation >

More happy Wallace-ness, using clusters this time to create the facial poses.

The primary visemes: Aah, Peh, and Eey (A,P,Y)

< Click here for the animation >

The point of this assignment was to create model poses (visemes) needed to make Wallace speak, and animate them with blendshapes.

The mouth poses used: three to form the word “happy”, and two others for refinement and a base pose.

A brief animation: here

showing a transition between happy & sad faces below and the neutral expression from Assignment 2: