Discussed Papers (2017 -)

Papers discussed listed in reverse chronological order.

May 2021 -

Data-Driven Approach to Simulating Realistic Human Joint Constraints

Anatomically Detailed Simulation of Human Torso

Data-Free Learning of Reduced-Order Kinematics (2)

Disentangling the Latent Space of 3D Human Body Meshes

Data-Free Learning of Reduced-Order Kinematics (1)

Learning Neural Constitutive Laws from Motion Observations for Generalizable PDE Dynamics

Addressing Discontinuous Root-Finding for Subsequent Differentiability in Machine Learning, Inverse Problems, and Control

Learning Neural Parametric Head Models

Fast and Feature-Complete Differentiable Physics for Articulated Rigid Bodies with Contact (2)

Fast and Feature-Complete Differentiable Physics for Articulated Rigid Bodies with Contact (1)

Generative GaitNet

UnderPressure: Deep Learning for Foot Contact Detection, Ground Reaction Force Estimation and Footskate Cleanup

COUCH: towards controllable human-chair interactions / Synthesizing Physical Character-Scene Interactions

Interactive Modelling of Volumetric Musculoskeletal Anatomy

Reconstructing Personalized Anatomical Models for Physics-based Body Animation (2)

Reconstructing Personalized Anatomical Models for Physics-based Body Animation (1)

Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations

Differentiable Simulation of Inertial Musculotendons

Smooth Non-Rigid Shape Matching via Effective Dirichlet Energy Optimization

NeuralSim: Augmenting differentiable simulators with neural networks (2)

NeuralSim: Augmenting differentiable simulators with neural networks (1)

Differentiable Dynamics for Articulated 3d Human Motion Reconstruction

Inferring Articulated Rigid Body Dynamics from RGBD Video

MeshCNN: A Network with an Edge

Physics-based Human Motion Estimation and Synthesis from Videos

Physics-based Deep Learning: Differentible Physics

Physics-based Deep Learning: Physical Losses

Physics-based Deep Learning: Intro

Graph Attention Networks

HodgeNet: Learning Spectral Geometry on Triangle Meshes

Attention Is All You Need

End-to-End Human Pose and Mesh Reconstruction with Transformers

Fast Quasi-Harmonic Weights for Geometric Data Interpolation (2)

Fast Quasi-Harmonic Weights for Geometric Data Interpolation (1)

DiffusionNet: Discretization Agnostic Learning on Surfaces (2)

DiffusionNet: Discretization Agnostic Learning on Surfaces (1)

Sept 2017 - March 2020

Geometric Deep Learning: Going beyond Euclidean data - Ch 5. Spectral Methods

Geometric Deep Learning: Going beyond Euclidean data - Ch 4. The Geometry of Manifolds and Graphs

Geometric Deep Learning: Going beyond Euclidean data - Ch 3. Deep Learning on Euclidean Domains

Geometric Deep Learning: Going beyond Euclidean data - Ch 1 & 2.

Poly-Spline Finite Element Method

Learning to Infer Implicit Surfaces without 3D Supervision

Coercing Machine Learning to Output Physically Accurate Results

Efficient and Accurate Collision Response for Elastically Deformable Models

Real2Sim: Visco-elastic parameter estimation from dynamic motion

Learning an Intrinsic Garment Space for Interactive Authoring of Garment Animation

Non-Smooth Newton Methods for Deformable Multi-Body Dynamics

A Scalable Galerkin Multigrid Method for Real-time Simulation of Deformable Objects

MeshCNN: A Network with an Edge

Subspace Neural Physics: Fast Data-Driven Interactive Simulation

Vibration-minimizing motion retargeting for robotic characters

Implicit Untangling: A Robust Solution for Modeling Layered Clothing

Numerical Coarsening using Discontinuous Shape Functions

DeepWarp: DNN-based Nonlinear Deformation

Analytic Eigensystems for Isotropic Distortion Energies

Anisotropic Elasticity for Inversion-Safety and Element Rehabilitation

Neural Material: Learning Elastic Constitutive Material and Damping Models from Sparse Data

All’s Well That Ends Well: Guaranteed Resolution of Simultaneous Rigid Body Impact

Latent-space Dynamics for Reduced Deformable Simulation

Fast Corotated FEM using Operator Splitting

DeepWrinkles: Accurate and Realistic Clothing Modeling

A Spectral BSSRDF for Shading Human Skin

Modeling and Estimation of Internal Friction in Cloth

A Hybrid Iterative Solver for Robustly Capturing Coulomb Friction in Hair Dynamics

A Nonsmooth Newton Solver for Capturing Exact Coulomb Friction in Fiber Assemblies

An Implicit Frictional Contact Solver for Adaptive Cloth Simulation

Proximal Algorithms (2)

Proximal Algorithms (1)

Flexible Simulation of Deformable Models Using Discontinuous Galerkin FEM

Quasi-Newton Methods for Real-time Simulation of Hyperelastic Materials