Differential Walk on Spheres
ACM Transactions on Graphics (SIGGRAPH Asia), 2024
We introduce a Monte Carlo method for computing derivatives of the solution to a partial differential equation (PDE) with respect to problem parameters (such as domain geometry or boundary conditions). Derivatives can be evaluated at arbitrary …
Walkin' Robin: Walk on Stars with Robin Boundary Conditions
ACM Transactions on Graphics (SIGGRAPH), 2024
🏆 Best Paper
Numerous scientific and engineering applications require solutions to boundary value problems (BVPs) involving elliptic partial differential equations, such as the Laplace or Poisson equations, on geometrically intricate domains. We develop a Monte …
fVDB: A Deep-Learning Framework for Sparse, Large-Scale, and High-Performance Spatial Intelligence
ACM Transactions on Graphics (SIGGRAPH), 2024
We present fVDB, a novel GPU-optimized framework for deep learning on large-scale 3D data. fVDB provides a complete set of differentiable primitives to build deep learning architectures for common tasks in 3D learning such as convolution, pooling, …
Monte Carlo Geometry Processing
Symposium on Geometry Processing Graduate School Courses, 2024
NeuralVDB: High-resolution Sparse Volume Representation using Hierarchical Neural Networks
ACM Transactions on Graphics (ToG), 2024
We introduce NeuralVDB, which improves on an existing industry standard for efficient storage of sparse volumetric data, denoted VDB [Museth 2013], by leveraging recent advancements in machine learning. Our novel hybrid data structure can reduce the …
Decorrelating ReSTIR Samplers via MCMC Mutations
ACM Transactions on Graphics (ToG), 2024
Monte Carlo rendering algorithms often utilize correlations between pixels to improve efficiency and enhance image quality. For real-time applications in particular, repeated reservoir resampling offers a powerful framework to reuse samples both …
XCube: Large-Scale 3D Generative Modeling using Sparse Voxel Hierarchies
ArXiv Preprint, 2023
We present X3 (pronounced XCube), a novel generative model for high-resolution sparse 3D voxel grids with arbitrary attributes. Our model can generate millions of voxels with a finest effective resolution of up to 10243 in a feed-forward fashion …
ViCMA: Visual Control of Multibody Animations
SIGGRAPH Asia (Conference Track), 2023
Motion control of large-scale, multibody physics animations with contact is difficult. Existing approaches, such as those based on optimization, are computationally daunting, and, as the number of interacting objects increases, can fail to find …
Subspace Mixed Finite Elements for Real-Time Heterogeneous Elastodynamics
SIGGRAPH Asia (Conference Track), 2023
Real-time elastodynamic solvers are well-suited for the rapid simulation of ho mogeneous elastic materials, with high-rates generally enabled by aggressive early termination of timestep solves. Unfortunately, the introduction of strong domain …
Neural Collision Fields for Triangle Primitives
SIGGRAPH Asia (Conference Track), 2023
We present neural collision fields as an alternative to contact point sampling in physics simulations. Our approach is built on top of a novel smoothed integral formulation for the contact surface patches between two triangle meshes. By reformulating …
Walk on Stars: Grid-Free Monte Carlo for Neumann Boundary Conditions
ACM Transactions on Graphics (SIGGRAPH), 2023
Grid-free Monte Carlo methods based on the walk on spheres (WoS) algorithm solve fundamental partial differential equations (PDEs) like the Poisson equation without discretizing the problem domain or approximating functions in a finite basis. Such …
Simulation-aided face strain extraction for ML animation systems
ACM SIGGRAPH 2023 Talks, 2023
We present a volumetric, simulation-based pipeline for the automatic creation of strain-based descriptors from facial performance capture provided as surface meshes. Strain descriptors encode facial poses via length elongation/contraction ratios of …
Interactive Hair Simulation on the GPU using ADMM
SIGGRAPH (Conference Track), 2023
We devise a local global solver dedicated to the simulation of Discrete Elastic Rods (DER) with Coulomb friction that can fully leverage the massively parallel compute capabilities of moderns GPUs. We verify that our simulator can reproduce …
Boundary Value Caching for Walk on Spheres
ACM Transactions on Graphics (SIGGRAPH), 2023
Grid-free Monte Carlo methods such as walk on spheres can be used to solve elliptic partial differential equations without mesh generation or global solves. However, such methods independently estimate the solution at every point, and hence do not …
Near-realtime Facial Animation by Deep 3D Simulation Super-Resolution
ArXiv Preprint, 2023
We present a neural network-based simulation super-resolution framework that can efficiently and realistically enhance a facial performance produced by a low-cost, realtime physics-based simulation to a level of detail that closely approximates that …
Deep Learned Face Swapping in Feature Film Production
ACM SIGGRAPH 2022 Talks, 2022
🏆 AIS Lumiere Tech Award
In visual effects for film, replacement of stunt performers’ facial likeness for their doubled actor counterparts using traditional computer graphics methods is a multi-stage, labor intensive task. Recently, deep learning techniques have made a …
An Exact Bitwise Reversible Integrator
ArXiv Preprint, 2022
At a fundamental level most physical equations are time reversible. In this paper we propose an integrator that preserves this property at the discrete computational level. Our simulations can be run forward and backwards and trace the same path …
NanoVDB: A GPU-Friendly and Portable VDB Data Structure For Real-Time Rendering And Simulation
ACM SIGGRAPH 2021 Talks, 2021
We introduce a sparse volumetric data structure, dubbed NanoVDB, which is portable to both C++11 and C99 as well as most graphics APIs, e.g. CUDA, OpenCL, OpenGL, WebGL, DirectX 12, OptiX, HLSL, and GLSL. As indicated by its name, NanoVDB is a …
Ultrahyperbolic Representation Learning
Advances in Neural Information Processing Systems (NeurIPS), 2020
In machine learning, data is usually represented in a (flat) Euclidean space where distances between points are along straight lines. Researchers have recently considered more exotic (non-Euclidean) Riemannian manifolds such as hyperbolic space which …
Computing Light Transport Gradients Using the Adjoint Method
ArXiv Preprint, 2020
This paper proposes a new equation from continuous adjoint theory to compute the gradient of quantities governed by the Transport Theory of light. Unlike discrete gradients ala autograd, which work at the code level, we first formulate the continuous …