Style-based Inverse Kinematics

Style-based Inverse Kinematics

Keith Grochow1     Steven L. Martin1     Aaron Hertzmann2     Zoran Popović1    

1University of Washington     2University of Toronto



Abstract
We present an inverse kinematics system based on a learned model of human poses. Given a set of constraints, our system can produce the most likely pose satisfying those constraints, in realtime. Training the model on different input data leads to different styles of IK. The model is represented as a probability distribution over the space of all possible poses. This means that our IK system can generate any pose, but prefers poses that are most similar to the space of poses in the training data. We represent the probability with a novel model called a Scaled Gaussian Process Latent Variable Model. The parameters of the model are all learned automatically; no manual tuning is required for the learning component of the system. We additionally describe a novel procedure for interpolating between styles.

Our style-based IK can replace conventional IK, wherever it is used in computer animation and computer vision. We demonstrate our system in the context of a number of applications: interactive character posing, trajectory keyframing, real-time motion capture with missing markers, and posing from a 2D image.

Citation
Keith Grochow, Steven L. Martin, Aaron Hertzmann, Zoran Popović. Style-based Inverse Kinematics. ACM Transactions on Graphics (Proceedings of SIGGRAPH 2004), 2004.

Paper
SIGGRAPH 2004 (1.4MB PDF)


Results

Examples of different styles
Baseball Pitch (2MB, Divx AVI)
Jump Shot (.7MB, Divx AVI)
Track Start (1MB, Divx AVI)
Posing characters
Posing with track style (9MB, Divx AVI)
Posing with track style using style window (6MB, Divx AVI)
Simple posing of a hand (4MB, Divx AVI)
Simulation of marker loss in real-time mocap
Video (14MB, Divx AVI)
Smooth intepolation between three walking styles
Video (11MB, Divx AVI)
Editing keyframed animation
Video (9MB, Divx AVI)
Posing a character using an image
Jumpshot (5MB, Divx AVI)
Baseball Pitch (4MB, Divx AVI)
Complete Siggraph video
High Res Version (720x480) (62MB, Divx AVI)
Low Res Version (360x240) (25MB, Divx AVI)


Errata
The right-hand-side of Equation 15 should be: K-1YW2YTK-1 - DK-1

Other Links
  • David MacKay's Introduction to Gaussian Processes
  • Neil Lawrence's page on Gaussian Process Latent Variable Model