← ArchivePaper2005
Creating and Simulating Skeletal Muscle from the Visible Human Data Set
Joseph Teran, Eftychios Sifakis, Silvia S. Blemker, Victor Ng-Thow-Hing, Cynthia Lau, Ronald Fedkiw
Finite element musculoskeletal simulation built from anatomical data, groundwork for the production flesh systems that followed.
Abstract
This paper presents a framework for extracting and simulating high resolution musculoskeletal geometry from the segmented visible human data set, demonstrated on roughly 30 contact coupled muscles of the upper limb made up of about 10 million tetrahedra. Muscle, tendon, and bone geometry is created using level set and constructive solid geometry repair, with B-spline solids assigning spatially varying fiber directions and a transversely isotropic, quasi-incompressible constitutive model providing active and passive fiber response. To make simulation tractable, each high resolution muscle is embedded in a nonmanifold, connectivity preserving simulation mesh molded from a lower resolution body-centered cubic lattice, which relaxes the time step restriction and reduces memory. A robust invertible finite element technique handles degenerate and inverted tetrahedra, and a fascia contact model maintains realistic contact between muscle groups during ballistic motion.
How to read this
- Category
- Method / system: FEM musculoskeletal modeling and simulation from anatomical data
- Contributions
- A framework to extract high-resolution muscle, tendon, and bone geometry from the segmented Visible Human data set using level sets and CSG repair, with B-spline solids assigning spatially varying fiber directions
- A transversely isotropic, quasi-incompressible constitutive model for active and passive fiber response, demonstrated on ~30 contact-coupled upper-limb muscles (~10 million tetrahedra)
- Embedding each high-res muscle in a connectivity-preserving BCC simulation mesh to relax the time-step restriction and cut memory, plus robust invertible FEM and a fascia contact model
- Context
- Grounds physically based flesh and muscle simulation in real anatomical data, providing the FEM musculoskeletal groundwork later built on by production flesh systems and by facial muscle work such as Sifakis et al. 2005.
- Correctness
- Anatomical fidelity is bounded by the Visible Human segmentation and the chosen constitutive model; the embedded lower-resolution simulation mesh trades some accuracy for tractability, and validation is by anatomical plausibility and stability rather than measured in-vivo mechanics.
- Clarity
- A substantial, technique-dense journal paper; a first pass conveys the geometry-extraction-to-simulation pipeline, with second and third passes needed for the constitutive model, invertible FEM, and embedding.
- How to read it
- Read for the end-to-end pipeline from anatomical data to simulable meshes; deep-dive the constitutive model, BCC embedding, and invertible FEM in later passes if you build muscle or flesh simulators.
Builds on
Nothing in the archive, this is a starting point.
Built upon by
- Automatic Determination of Facial Muscle Activations from Sparse Motion Capture Marker Data 2005
- Robust Quasistatic Finite Elements and Flesh Simulation 2005
- Musculotendon Simulation for Hand Animation 2008
- Physics-based Character Skinning using Multi-Domain Subspace Deformations 2011
- Anatomy Transfer 2013
- Capture and Statistical Modeling of Arm-Muscle Deformations 2013
- Active Volumetric Musculoskeletal Systems 2014
- Biomechanical Simulation and Control of Hands and Tendinous Systems 2015
- Flesh, Flab, and Fascia Simulation on Zootopia 2016
- How to Build a Human: Practical Physics-Based Character Animation 2016
- Shape Targeting: A Versatile Active Elasticity Constitutive Model 2020
- Creating a Photorealistic Hyena 2022
- Loki: A Unified Multiphysics Simulation Framework for Production 2022
- The Making of CG Animals 2023
- CFX: Muscle and Soft Tissue with Otis | Houdini 21 | Kai Stavginski | SIGGRAPH HIVE 2025 2025
Related work
- How to Build a Human: Practical Physics-Based Character Animation 2016 / DigiPro
- Robust Quasistatic Finite Elements and Flesh Simulation 2005 / SCA
- Lessons from the Evolution of an Anatomical Facial Muscle Model 2017 / DigiPro
- A Neural Network Model for Efficient Musculoskeletal-Driven Skin Deformation 2024 / SIGGRAPH
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 →