Self-supervised Learning of Latent Space Dynamics
Authors:
- Yue LiYLYue Li
- Gene Wei-Chin LinGLGene Wei-Chin Lin
- Egor LarionovELEgor Larionov
- Aljaz BozicABAljaz Bozic
- Doug RobleDRDoug Roble
- Ladislav KavanLKLadislav Kavan
- Stelian CorosSCStelian Coros
- Bernhard ThomaszewskiBTBernhard Thomaszewski
- Tuur StuyckTSTuur Stuyck
- Hsiao-Yu ChenHCHsiao-Yu Chen
Preprint
Abstract
Modeling the dynamic behavior of deformable objects is crucial for creating realistic digital worlds. While conventional simulations produce high-quality motions, their computational costs are often prohibitive. Subspace simulation techniques address this challenge by restricting deformations to a lower-dimensional space, improving performance while maintaining visually compelling results. However, even subspace methods struggle to meet the stringent performance demands of portable devices such as virtual reality headsets and mobile platforms. To overcome this limitation, we introduce a novel subspace simulation framework powered by a neural latent-space integrator. Our approach leverages self-supervised learning to enhance inference stability and generalization. By operating entirely within latent space, our method eliminates the need for full-space computations, resulting in a highly efficient method well-suited for deployment on portable devices. We demonstrate the effectiveness of our approach on challenging examples involving rods, shells, and solids, showcasing its versatility and potential for widespread adoption.Keywords
All Related Versions
This publication has 23 references indexed in Scilit:
- Real-Time Hair Simulation With Neural InterpolationIEEE Transactions on Visualization and Computer Graphics, 2020
- Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencodersJournal of Computational Physics, 2020
- A reduced model for interactive hairsACM Transactions on Graphics, 2014
- Subspace clothing simulation using adaptive basesACM Transactions on Graphics, 2014
- Subspace integration with local deformationsACM Transactions on Graphics, 2013
- Subspace fluid re-simulationACM Transactions on Graphics, 2013
- FEM simulation of 3D deformable solids2012
- Hair meshesACM Transactions on Graphics, 2009
- Model reduction for real-time fluidsACM Transactions on Graphics, 2006
- Real-Time subspace integration for St. Venant-Kirchhoff deformable modelsACM Transactions on Graphics, 2005