Parametric Skeletons with Reduced Soft‐Tissue Deformations

Abstract

We present a method to augment parametric skeletal models with subspace soft-tissue deformations. We combine the benefits of data-driven skeletal models, i.e. accurate replication of contact-free static deformations, with the benefits of pure physics-based models, i.e. skin and skeletal reaction to contact and inertial motion with two-way coupling. We succeed to do so in a highly efficient manner, thanks to a careful choice of reduced model for the subspace deformation. With our method, it is easy to design expressive reduced models with efficient yet accurate force computations, without the need for training deformation examples. We demonstrate the application of our method to parametric models of human bodies, SMPL, and hands, MANO, with interactive simulations of contact with nonlinear soft-tissue deformation and skeletal response.>

Learning contact corrections for handle-based subspace dynamics

“…Linear subspace models have also been used to simulate only local deformations on top of some different global deformation model [Kim and James 2011]. Moreover, subspace local deformations can be aggregated with other deformations, such as pose-based blendshapes [Tapia et al 2021]. As a way to increase the accuracy of subspace deformations, several works [Hahn et al 2014;Xu and Barbič 2016] blend local linear subspace models in a pose-dependent manner.…”

Survey on Hand-Based Haptic Interaction for Virtual Reality

“…There are two constraints to achieving the above goals, one is the high complexity of virtual hand modeling, and the other is the limitation of hand tracking and mapping, such as high-DoF motion matching between the real hand and virtual hand, especially for the thumb and palm, and the adaptability to different hand sizes (e.g., width, length, and circumference). Data-driven methods possess advantages of high efficiency and accuracy for contactfree static poses, and subspace simulation methods provide a fast way to approximate soft-tissue deformations [131], [132]. A hybrid of such two approaches has the potential to enhance the realism of hand-based interactions.…”

FLSH - Friendly Library for the Simulation of Humans

“…the vector of reduced DoFs and > the matrix of basis coefficients. We use as reduced model the bounded generalized biharmonic coordinates of [7], following the model architecture of [6]. This allows seamless integration in the basis of handle points in the skin together with the skeletal bones.…”

“…Using a reduced basis instead of a full FEM discretization dramatically reduces the solver cost, thanks to the reduced size of the problem gradient and Hessian. We also add cubature, following again the data-oblivious approach of [6], to approximate the skin inertia and elastic energy.…”

“…ROM soft skeleton: This is a reduced-order model (ROM) of the FEM soft skeleton. Bodies solve a real-time approximation of soft-tissue deformation, including contact, following a research solution [6]. This representation offers an optimal trade-off between accuracy and performance.…”