← ArchivePaper2018
Smeat: ADMM Based Tools for Character Deformation
ADMM-based creature tools for muscle, skin, and fatty tissue simulation achieving anatomically accurate dynamic deformation, demonstrated on Avengers: Infinity War characters.
Abstract
Recent work on physical simulation in computer graphics has focused on energy minimization formulations of dynamics based on fast optimization methods. Because these methods can efficiently tackle a wide range of problems in the realms of both dynamic simulation and static relaxation, we adopted one such method, ADMM, and used it to implement a set of creature tools we call Smeat. We will describe this tool set and give a case study of how it was used to create character effects on "Avengers: Infinity War".
How to read this
- Category
- Production talk / tools (ADMM-based creature deformation)
- Contributions
- Adopts ADMM optimization to build a creature tool set (Smeat) spanning dynamic simulation and static relaxation
- Simulates muscle, skin, and fatty tissue for anatomically driven dynamic deformation
- Presents a case study of creating character effects on Avengers: Infinity War
- Context
- Relates to energy-minimization / fast-optimization approaches to physical simulation (in the lineage of Projective Dynamics), applying the ADMM solver to muscle, skin, and fat tissue.Builds on: Projective Dynamics: Fusing Constraint Projections for Fast Simulation
- Correctness
- Studio practice, not peer-reviewed; results are production-proven via the Infinity War case study, so the anatomical accuracy claim should be read as art-directed plausibility rather than measured biomechanics.
- Clarity
- Accessible as a tools-and-case-study briefing; a first pass conveys what the tool set does, with the ADMM formulation only sketched.
- How to read it
- Read once for the tool scope (muscle/skin/fat) and why ADMM suits both dynamics and static relaxation; consult the underlying ADMM/Projective Dynamics literature separately if you need the math.
Built upon by
Nothing yet.
Related work
- Muscle and Fascia Simulation with Extended Position Based Dynamics 2020 / CGF
- Clean Cloth Inputs: Removing Character Self-Intersections with Volume Simulation 2018 / SIGGRAPH
- Finding Hank 2016 / SIGGRAPH
- Efficient and Robust Skin Slide Simulation 2017 / DigiPro
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 →