Efficient dynamic skinning with low-rank helper bone controllers

Mukai, Tomohiko; Kuriyama, Shigeru

doi:10.1145/2897824.2925905

Cited by 30 publications

(17 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Furthermore, an animator would need to learn to use the fitted bone structure in order to author new animations. Example-based skinning methods have been developed [Feng et al 2008;Mohr and Gleicher 2003;Mukai and Kuriyama 2016;Wang et al 2007] that uses the original skeleton from a character rig. These methods use training examples to learn a deformation model that can approximate mesh deformations given new skeleton poses not seen during model training.…”

Section: Related Workmentioning

confidence: 99%

Fast and deep deformation approximations

et al. 2018

View full text Add to dashboard Cite

Character rigs are procedural systems that compute the shape of an animated character for a given pose. They can be highly complex and must account for bulges, wrinkles, and other aspects of a character's appearance. When comparing film-quality character rigs with those designed for real-time applications, there is typically a substantial and readily apparent difference in the quality of the mesh deformations. Real-time rigs are limited by a computational budget and often trade realism for performance. Rigs for film do not have this same limitation, and character riggers can make the rig as complicated as necessary to achieve realistic deformations. However, increasing the rig complexity slows rig evaluation, and the animators working with it can become less efficient and may experience frustration. In this paper, we present a method to reduce the time required to compute mesh deformations for film-quality rigs, allowing better interactivity during animation authoring and use in real-time games and applications. Our approach learns the deformations from an existing rig by splitting the mesh deformation into linear and nonlinear portions. The linear deformations are computed directly from the transformations of the rig's underlying skeleton. We use deep learning methods to approximate the remaining nonlinear portion. In the examples we show from production rigs used to animate lead characters, our approach reduces the computational time spent on evaluating deformations by a factor of 5X-10X. This significant savings allows us to run the complex, film-quality rigs in real-time even when using a CPU-only implementation on a mobile device.

show abstract

Section: Related Workmentioning

confidence: 99%

Fast and deep deformation approximations

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Loper et al [17] showed how realistic pose and body shapes can be extracted from standard motion capture markers by using a data-driven human body model, and Tsoli et al [18] focused on capture-based modeling of breathing. Fast models for real-time applications rely on combining linear blend shapes [19], rotation regression [6] or helper-bone rigs that have proven useful for delivering effects such as muscle bulging [5,20]. Another exciting strand of current research is learning dynamics from data, either using statistical methods [21] or by combining statistical methods with physics-based simulation [22].The most recent work [22] is starting to show promise of being able to generalize to imaginary artist-designed characters.…”

Section: Related Workmentioning

confidence: 99%

Dynamic implicit muscles for character skinning

Roussellet

Rumman

Canezin

et al. 2018

Computers & Graphics

View full text Add to dashboard Cite

Most current methods for character skinning can be categorized into 1) geometric techniques, which are fast and easy to use but often lack physical realism, 2) data-driven approaches, which require a large set of examples that are tedious to edit, and 3) physics-based methods, which are highly realistic but slow and difficult to use. Recently introduced geometric Implicit Skinning methods can solve contact interactions and skin elasticity with results comparable to physics-based simulation in real-time. In this paper we introduce an animation method that adds anatomical plausibility while benefiting from the advantages of Implicit Skinning. We propose an efficient way to model muscle primitives with implicit surfaces. Volumetric extrusions of individual muscles are attached to muscle center lines simulated with a fast, low-dimensional physics-based approach (Position Based Dynamics of one-dimensional line segments). This combination of physics-based simulation with implicit modeling allows us to elegantly resolve muscle-muscle and muscle-bone collisions and add dynamic effects such as flesh jiggling while guaranteeing volume preservation (which is a property of real biological muscles) and producing visually plausible skin-skin contact behavior. Our method runs at interactive frame-rates and features intuitive modeling parameters which allow animators to quickly explore a variety of designs and physics-based effects.

show abstract

“…Many techniques have been proposed to overcome this problem [McAdams et al 2011]. Larboulette et al [2005] used a simple blend between static skinning and dynamic deformation, and Mukai and Kuriyama [2016] added helper bones that add dynamic behaviors to LBS. The dimensionality reduction is an effective and principled approach for efficient simulation of deformable bodies.…”

Section: Related Workmentioning

confidence: 99%

Data-driven physics for human soft tissue animation

Kim

Pons-Moll²,

Pujades³

et al. 2017

ACM Trans. Graph.

View full text Add to dashboard Cite

Data driven models of human poses and soft-tissue deformations can produce very realistic results, but they only model the visible surface of the human body and cannot create skin deformation due to interactions with the environment. Physical simulations can generalize to external forces, but their parameters are difficult to control. In this paper, we present a layered volumetric human body model learned from data. Our model is composed of a data-driven inner layer and a physics-based external layer. The inner layer is driven with a volumetric statistical body model (VSMPL). The soft tissue layer consists of a tetrahedral mesh that is driven using the finite element method (FEM). Model parameters, namely the segmentation of the body into layers and the soft tissue elasticity, are learned directly from 4D registrations of humans exhibiting soft tissue deformations. The learned two layer model is a realistic full-body avatar that generalizes to novel motions and external forces. Experiments show that the resulting avatars produce realistic results on held out sequences and react to external forces. Moreover, the model supports the retargeting of physical properties from one avatar when they share the same topology.

show abstract

Efficient dynamic skinning with low-rank helper bone controllers

Cited by 30 publications

References 33 publications

Fast and deep deformation approximations

Fast and deep deformation approximations

Dynamic implicit muscles for character skinning

Data-driven physics for human soft tissue animation

Contact Info

Product

Resources

About