Discover and read the best of Twitter Threads about #siggraph2022
Most recents (5)
Here's a quick @beforesmag THREAD on some of the special presentations studios and companies are doing at @siggraph #siggraph2022 in Vancouver... 
First up, it's the @autodesk Vision Series (I'm particularly looking forward to 'Camila Cabello & the Toughest Green Screen Shot I’ve Ever Undertaken')
cvent.autodesk.com/event/bcf9e9f7…
cvent.autodesk.com/event/bcf9e9f7…
Then, @EpicGames and @UnrealEngine are putting on a whole bunch of sessions (be great to see where things like Lumen are up to) unrealengine.com/en-US/events/j…
Models in engineering & science have *way* more complexity in geometry/materials than what conventional solvers can handle.
But imagine if simulation was like Monte Carlo rendering: just load up a complex model and hit "go"; don't worry about meshing, basis functions, etc. [1/n]
But imagine if simulation was like Monte Carlo rendering: just load up a complex model and hit "go"; don't worry about meshing, basis functions, etc. [1/n]
Our #SIGGRAPH2022 paper takes a major step toward this vision by building a bridge between PDEs & volume rendering: cs.dartmouth.edu/wjarosz/public…
Joint work with
@daseyb — darioseyb.com
@rohansawhney1 — rohansawhney.io
@wkjarosz — cs.dartmouth.edu/wjarosz/
[2/n]
Joint work with
@daseyb — darioseyb.com
@rohansawhney1 — rohansawhney.io
@wkjarosz — cs.dartmouth.edu/wjarosz/
[2/n]
Here's one fun example: heat radiating off of infinitely many aperiodically-arranged black body emitters, each with super-detailed geometry, and super-detailed material coefficients.
From this view alone, the *boundary* meshes have ~600M vertices. Try doing that with FEM! [3/n]
From this view alone, the *boundary* meshes have ~600M vertices. Try doing that with FEM! [3/n]
Introducing DeltaConv: an INSANE convolution layer for point clouds. It's INtrinsic, Surface-first, ANisotropic, and... erm... Easy to use! Coming to #SIGGRAPH2022 💾code 📜paper and 🧙summary at rubenwiersma.nl/deltaconv 1/8
Anisotropic convolution is a central building block of CNNs, but challenging to transfer to surfaces, because there's no global coordinate system to align our filters. 2/8
The missing global coordinate system is addressed in differential geometry by using coordinate-independent operators, such as the Laplacian ❤️ but a downside of only using the Laplacian is that it is isotropic 💔 3/8
Thrilled to share "GANimator: Neural Motion Synthesis from a Single Sequence", in this upcoming #SIGGRAPH2022 🤩
GANimator can produce novel animations for unique creatures that don't have large motion datasets! For example, this hexapedal crab 🦀(1/4)
GANimator can produce novel animations for unique creatures that don't have large motion datasets! For example, this hexapedal crab 🦀(1/4)
We are inspired by the seminal SinGAN paper 🚀 We only use the information within a single motion sequence! GANimator is a unified framework which addresses real applications in animation: eg crowd simulation, key-frame editing, style transfer, & interactive control (2/4)
GANimator produces both conditional & unconditional sequences that are reminiscent of the input motion. We generate sequences that still resemble the core elements of the original animation, while simultaneously synthesizing novel and diverse movements 🕺💃👯(3/4)
Super happy to share our @siggraph paper with @nvidia "MatBuilder: Mastering Sampling Uniformity Over Projections" (@LoisResearch, @nbonneel, @dcoeurjo, JC.Iehl, A.Keller, V.Ostromoukhov) projet.liris.cnrs.fr/matbuilder/. The motivation is that to estimate the value of an integral [1/16]
we can randomly sample the integrand and sum values -- this is Monte Carlo integration. With "well spread" samples instead of naive random values, we can improve the accuracy of the integral estimate a lot -- this is the quasi-Monte Carlo method. [2/16]
For instance, we can use stratified samples: for example, if we want 16 samples to estimates a 2-d integral defined in the domain [0,1]x[0,1], we decompose the domain into 4x4 square cells, and we put one sample in each cell. [3/16]
