← ArchivePaper2017
Enriching Facial Blendshape Rigs with Physical Simulation
Yeara Kozlov, Derek Bradley, Moritz Bächer, Bernhard Thomaszewski, Thabo Beeler, Markus Gross
Augments blendshape rigs with finite-element tissue simulation to add physically plausible secondary dynamics during facial animation.
Abstract
This work presents a method for adding physical effects such as secondary dynamics to traditional facial blendshape animation by enriching blendshape rigs with a simple volumetric tissue structure. Rather than relying on detailed anatomical models, the approach uses the artist-created blendshape animation as per-frame rest shapes for a finite element simulation, ensuring that in the absence of head motion or external forces the result exactly matches the input animation. The method introduces blendmaterials to account for spatio-temporally varying material properties caused by muscle activation, and a dynamic rebalancing scheme that updates the deformed configuration to preserve inertial forces and avoid spurious oscillations when changing rest shape or material. Results are demonstrated on both realistic human faces and fantasy creatures undergoing rapid head motion, running, and speech.
How to read this
- Category
- Method: physics enrichment of facial blendshape rigs
- Contributions
- A method that adds secondary dynamics to blendshape animation by enriching the rig with a simple volumetric tissue structure and using artist animation as per-frame FEM rest shapes
- Blendmaterials, modeling spatio-temporally varying material properties caused by muscle activation
- A dynamic rebalancing scheme that updates the deformed configuration to preserve inertial forces and avoid spurious oscillations when rest shape or material changes
- Context
- Extends artist-directed facial rigging in the lineage of Bickel et al.'s Physically Based Rigging for Artist-Directed Facial Animation, layering FEM dynamics onto standard blendshapes without a detailed anatomical model.Builds on: Physically Based Rigging for Artist-Directed Facial Animation
- Correctness
- Demonstrated on realistic human faces and fantasy creatures under rapid head motion, running, and speech; by construction it matches the input animation when there is no head motion or external force, so the contribution is the added dynamics, and fidelity depends on the simple tissue model rather than true anatomy.
- Clarity
- Technical but well-motivated; a first pass conveys the rest-shape-from-animation idea, a second pass is needed for blendmaterials and the rebalancing scheme.
- How to read it
- Focus on how artist blendshapes become per-frame rest shapes and why rebalancing is needed; a second pass pays off for the blendmaterial formulation if you implement secondary facial dynamics.
Built upon by
Related work
- Lessons from the Evolution of an Anatomical Facial Muscle Model 2017 / DigiPro
- Fully Automatic Generation of Anatomical Face Simulation Models 2015 / SCA
- Art-Directed Muscle Simulation for High-End Facial Animation 2016 / SCA
- Realtime Performance-Driven Physical Simulation for Facial Animation 2019 / CGF
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 →