← ArchivePaper2025
BlendSim: Simulation on Parametric Blendshapes using Spacetime Projective Dynamics
BlendSim shifts physics-based animation from discrete frame-by-frame solves to continuous interpolation trajectories built on blendshapes, representing each vertex path with continuous parametric Bezi
Abstract
BlendSim shifts physics-based animation from discrete frame-by-frame solves to continuous interpolation trajectories built on blendshapes, representing each vertex path with continuous parametric Bezier splines that have variable keyframe times. Because this mesh animation representation is continuous and fully differentiable, it can be optimized to follow the laws of physics under constraints, while projective dynamics decouples the optimization into local parallelizable steps and a global quadratic step for efficiency. The method supports constraints such as collisions and cyclic motion and is compatible with modern animation workflows and file formats such as glTF.
How to read this
- Category
- Method: physics-based animation
- Contributions
- BlendSim, shifting physics-based animation from discrete frame-by-frame solves to continuous interpolation trajectories built on blendshapes
- Represents each vertex path with continuous, differentiable parametric Bezier splines with variable keyframe times, optimizable to follow physics under constraints
- Uses projective dynamics to decouple the solve into parallelizable local steps and a global quadratic step, supporting collisions and cyclic motion and exporting to glTF
- Context
- Built on Projective Dynamics, recasting that fast simulation framework in a continuous, blendshape-and-spline trajectory representation rather than per-frame state.Builds on: Projective Dynamics: Fusing Constraint Projections for Fast Simulation
- Correctness
- The key idea is that a continuous, differentiable spline-on-blendshapes representation can be optimized to satisfy physical laws and constraints; readers should keep in mind that continuous trajectory parameterization trades temporal resolution and keyframe placement against the range of dynamics it can faithfully represent, and check which constraint types and motions were demonstrated.
- Clarity
- Conceptually crisp framing (continuous vs discrete), but the spacetime projective-dynamics formulation warrants a careful second pass.
- How to read it
- First pass for the continuous-trajectory reframing and why differentiability matters; second pass on the spacetime projective-dynamics local/global decomposition if you intend to implement, and note the glTF and animation-workflow compatibility as a practical hook.
Built upon by
Nothing yet.
Related work
- An Implicit Physical Face Model Driven by Expression and Style 2023 / SIGGRAPH Asia
- Gravity Preloading for Maintaining Hair Shape Using the Simulator as a Closed-Box Function 2022 / SIGGRAPH
- Clean Cloth Inputs: Removing Character Self-Intersections with Volume Simulation 2018 / SIGGRAPH
- Hyper-Reduced Projective Dynamics 2018 / TOG
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 →