← ArchivePaper2006

Inverse Kinematics for Reduced Deformable Models

Kevin G. Der, Robert W. Sumner, Jovan Popovic

Automated pipeline converts unarticulated example shapes into a controllable articulated reduced-deformable model with intuitive IK-style control.

Open the paper ↗Trace in the lineage graph

Abstract

Articulated shapes are aptly described by reduced deformable models that express required shape deformations using a compact set of control parameters. Although sufficient to describe most shape deformations, these control parameters can be ill-suited for animation tasks, particularly when reduced deformable models are inferred automatically from example shapes. Our algorithm provides intuitive and direct control of reduced deformable models similar to a conventional inverse-kinematics algorithm for jointed rigid structures. We present a fully automated pipeline that transforms a set of unarticulated example shapes into a controllable, articulated model. With only a few manipulations, an animator can automatically and interactively pose detailed shapes at rates independent of their geometric complexity.

How to read this

Category: Method: IK-style control for reduced deformable models
Contributions: A fully automated pipeline that turns a set of unarticulated example shapes into a controllable, articulated reduced-deformable model.
An inverse-kinematics-style interface that gives intuitive, direct control over the model's compact parameters.
Interactive posing of detailed shapes at rates independent of geometric complexity.
Context: Extends example-based deformation control in the spirit of Sumner et al.'s Mesh-Based Inverse Kinematics, bringing jointed-rigid IK intuition to automatically inferred reduced deformable models.Builds on: Mesh-Based Inverse Kinematics
Correctness: Demonstrated as an interactive posing tool over example shapes, with control rates decoupled from mesh complexity; the quality of poses is bounded by how well the example set spans the desired deformations, so coverage of the examples is the key assumption to keep in mind.
Clarity: Accessible at a high level; a first pass conveys the IK-from-examples idea, a second pass is needed for the reduced-model formulation and the control mapping.
How to read it: First pass for the pipeline (examples to articulated model to IK control); do a second pass on how control parameters are derived and solved if you plan to implement or compare against it.

Builds on

Mesh-Based Inverse Kinematics 2005

Built upon by

Nothing yet.

Related work

Mesh-Based Inverse Kinematics 2005 / SIGGRAPH
Easy Rigging of Face by Automatic Registration and Transfer of Skinning Parameters 2010 / ICCVG
S3: Neural Shape, Skeleton, and Skinning Fields for 3D Human Modeling 2021 / CVPR
Mobilizing Mocap, Motion Blending, and Mayhem: Rig Interoperability for Crowd Simulation on Incredibles 2 2018 / SIGGRAPH

Keywords

inverse kinematics reduced deformable models deformation gradients direct manipulation mesh posing proxy vertices skinning weights affine control transformations nonlinear optimization gauss-newton cholesky factorization resolution-independent mesh-based inverse kinematics shape interpolation example shapes skeletal deformation

This page summarises the entry and links to its original source. The archive never hosts or redistributes the publication itself.Show it in the full archive list →