Latent-space Dynamics for Reduced Deformable Simulation

dc.contributor.authorFulton, Lawsonen_US
dc.contributor.authorModi, Vismayen_US
dc.contributor.authorDuvenaud, Daviden_US
dc.contributor.authorLevin, David I. W.en_US
dc.contributor.authorJacobson, Alecen_US
dc.contributor.editorAlliez, Pierre and Pellacini, Fabioen_US
dc.date.accessioned2019-05-05T17:41:37Z
dc.date.available2019-05-05T17:41:37Z
dc.date.issued2019
dc.description.abstractWe propose the first reduced model simulation framework for deformable solid dynamics using autoencoder neural networks. We provide a data-driven approach to generating nonlinear reduced spaces for deformation dynamics. In contrast to previous methods using machine learning which accelerate simulation by approximating the time-stepping function, we solve the true equations of motion in the latent-space using a variational formulation of implicit integration. Our approach produces drastically smaller reduced spaces than conventional linear model reduction, improving performance and robustness. Furthermore, our method works well with existing force-approximation cubature methods.en_US
dc.description.number2
dc.description.sectionheadersLearning to Animate
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume38
dc.identifier.doi10.1111/cgf.13645
dc.identifier.issn1467-8659
dc.identifier.pages379-391
dc.identifier.urihttps://doi.org/10.1111/cgf.13645
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf13645
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectComputing methodologies
dc.subjectPhysical simulation
dc.subjectDimensionality reduction and manifold learning
dc.titleLatent-space Dynamics for Reduced Deformable Simulationen_US
Files
Original bundle
Now showing 1 - 2 of 2
Loading...
Thumbnail Image
Name:
v38i2pp379-391.pdf
Size:
20.77 MB
Format:
Adobe Portable Document Format
No Thumbnail Available
Name:
main_final-handbrake.mp4
Size:
18.61 MB
Format:
Unknown data format
Collections