Face Robot is the culmination of many years of animation research and development at Softimage, which has been at the forefront of computer graphics technology for nearly 20 years, spearheading many fundamental animation technologies, such as inverse kinematics, which first found their application in Softimage’s 3D systems and enabled ground-breaking new results, such as the digital dinosaurs in JURASSIC PARK.

The technology for Face Robot was developed in secrecy by the Softimage Special Projects team and condenses numerous algorithmic advances and years of production experience with facial animation into the first coherent solution for facial animation. Led by Isner, Thomas Kang and Javier von der Pahlen of the special projects team, Face Robot was developed in close collaboration with one of the key partners in this project, Jeff Wilson at Blur Studio in Los Angeles, who helped define the artistic needs for creating life-like facial animation.

Essentially, there are six stages to Face Robot involved in preparing a human face for animation:

Stage 1—Assembly: Brings together facial parts and check their validity.
* Import a completed 3D model of a head built according to Face Robot modeling guidelines.
* Check if meshes are ready for Face Robot.
* Fix holes.
* Generate the interior of the eyes and mouth.

Stage 2 —Pick Objects: identify the mesh components of the face.
* Pick the face mesh, eyes and teeth.

Stage 3 —Pick Landmarks: select landmark points on the face.
* Visually guided workflow to pick the key points on the face.

Stage 4 — Fit: spatially aligns specialized parts of the face.
* Fit jaw tissue, neck and jawbone.

Stage 5 —Act by setting keyframes or by applying motion capture.
* Key facial animation controls.
* Load facial motion capture from .c3d files.
* Re-author and alter performance using motion retargeting.
* Blend keyframes and motion capture animation.
* Use the animation mixer to blend animation.
* Animate eye blinking.
* Test facial poses.
* Export animation to XSI/Maya/Max/ LightWave/Messiah.

Stage 6 —Tune: adjust the mechanical behavior of the face to match artistic or data-driven goals. This last process step consists of soft tissue tuning, an iterative technique that uses various forms of manipulation of the soft tissue model to achieve the desired deformation of the face’s surface for a range of facial expressions.
* Adjust the soft tissue regions of the face through sculpting.
* Adjust the eyelids.
* Adjust the mouth.
* Create maps to control facial behavior through rendering or to transfer to a game engine.

Softimage Co. (www.softimage.com), a subsidiary of Avid Technology Inc., delivers innovative, artist-friendly character creation and effects tools to animators and digital artists in the film, broadcast, post-production and games industries. Its flagship product, SOFTIMAGE|XSI — the industry's only non-destructive digital character production software — frees character animators and digital artists to create compelling animated 3D content and effects: from major motion pictures to CG features and commercials, to videogames.

Avid Technology Inc. (www.avid.com) is a leader in digital nonlinear media creation, management and distribution solutions, enabling film, video, audio, animation, games and broadcast professionals to work more efficiently, productively, and creatively.