← ArchivePaper2016
Pose-Space Subspace Dynamics
Multiple FEM subspace models are precomputed around representative poses and blended at runtime, adding millisecond-cost secondary dynamics to rigged characters.
Abstract
We enrich character animations with secondary soft-tissue Finite Element Method (FEM) dynamics computed under arbitrary rigged or skeletal motion. Our method optionally incorporates pose-space deformation (PSD). It runs at milliseconds per frame for complex characters, and fits directly into standard character animation pipelines. Our simulation method does not require any skin data capture; hence, it can be applied to humans, animals, and arbitrary (real-world or fictional) characters. In standard model reduction of three-dimensional nonlinear solid elastic models, one builds a reduced model around a single pose, typically the rest configuration. We demonstrate how to perform multi-model reduction of Finite Element Method (FEM) nonlinear elasticity, where separate reduced models are precomputed around a representative set of object poses, and then combined at runtime into a single fast dynamic system, using subspace interpolation. While time-varying reduction has been demonstrated before for offline applications, our method is fast and suitable for hard real-time applications in games and virtual reality. Our method supports self-contact, which we achieve by computing linear modes and derivatives under contact constraints.
How to read this
- Category
- Method: real-time secondary soft-tissue dynamics for rigged characters
- Contributions
- Adds secondary FEM soft-tissue dynamics under arbitrary rigged or skeletal motion at milliseconds per frame, fitting standard animation pipelines
- Multi-model reduction: separate reduced FEM models precomputed around representative poses and combined at runtime via subspace interpolation
- Self-contact support via linear modes, and optional integration with pose-space deformation, requiring no skin capture so it applies to humans, animals, and fictional characters
- Context
- Builds on pose-space deformation in the lineage of Lewis et al.'s PSD, extending single-pose model reduction to multiple pose-anchored reduced models.Builds on: Pose Space Deformation: A Unified Approach to Shape Interpolation and Skeleton-Driven Deformation
- Correctness
- Targets hard real-time use; the key trade-off is that fidelity depends on the chosen representative pose set and on the reduced-subspace size, so motions far from the sampled poses are approximated by interpolation.
- Clarity
- Moderately technical; a first pass conveys the multi-model-reduction idea, but the subspace interpolation and contact handling reward a second pass.
- How to read it
- First pass for the multi-pose reduced-model concept and where it slots into a rig; second pass on the subspace interpolation and self-contact modes if implementing real-time secondary dynamics.
Builds on
Built upon by
Nothing yet.
Related work
- CUDA Deformers for Model Reduction 2020 / MIG
- Simulation of Hand Anatomy Using Medical Imaging 2022 / SIGGRAPH Asia
- Complementary Dynamics 2020 / SIGGRAPH Asia
- A Deep Emulator for Secondary Motion of 3D Characters 2021 / CVPR
Keywords
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 →