Paul C. DiLorenzo

Laughing out loud: control for modeling anatomically inspired laughter using audio

We present a novel technique for generating animation of laughter for a character. Our approach utilizes an anatomically inspired, physics-based model of a human torso that includes a mix of rigid-body and deformable components and is driven by Hill-type muscles. We propose a hierarchical control method which synthesizes laughter from a simple set of input signals. In addition, we present a method for automatically creating an animation from a soundtrack of an individual laughing. We show examples of laugh animations generated by hand-selected input parameters and by our audio-driven optimization approach. We also include results for other behaviors, such as coughing and a sneeze, created using the same model. These animations demonstrate the range of possible motions that can be generated using the proposed system. We compare our technique with both data-driven and procedural animations of laughter.

Breathe easy: model and control of human respiration for computer animation

In this paper, we detail an anatomically inspired, physically based model of the human torso designed for the visual simulation of respiration using a mixed system of rigid and deformable parts. Motion related to breath is a signature movement of the human body and an indicator for life but it has been largely overlooked by the graphics community. A novel composition of biological components is necessary to capture the key characteristics of breathing motion visible in the human trunk because the movement is generated fundamentally through the combination of both rigid bone and soft tissue. Our approach uses a simple physically based muscle element which is used throughout to drive the motion of the ribs and diaphragm as well as in other muscles, like those of the abdomen, to produce passive resistance. In addition, we describe an implementation of a straightforward method for preserving incompressible volume in deformable bodies to use in approximating the motion of the abdomen related to breath. Through the careful construction of this anatomically based torso, control for respiration becomes the generation of periodic contraction signals for a minimal set of two muscle groups. We show the flexibility of our approach through the animation of several breathing styles using our system.

Laughing out loud

We present a novel technique to generate and control laughter using a physically based, anatomically inspired torso simulation. We use this system to synthesize motion of the trunk and to create secondary effects that propagate to other parts of the body, e.g. the arms and the head. A hybrid set of rigid and flexible components comprise the model of the torso: spine, ribs, diaphragm, and abdomen as well as neck and shoulders. We employ hierarchical, Hill-type muscles to actuate laughter movement. The result is a rich, controlled motion derived from a simple, intuitive set of control parameters which we can use to demonstrate a range of laughing animations.

Premo: a natural-interaction animation platform

classroom use is granted without fee provided that copies are not made or distributed for commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of this work must be honored. For all other uses, contact the Owner/Author. SIGGRAPH 2014, August 10 – 14, 2014, Vancouver, British Columbia, Canada. 2014 Copyright held by the Owner/Author. ACM 978-1-4503-2960-6/14/08 Premo: A Natural-Interaction Animation Platform

Model and control of simulated respiration for animation

Animation of the breath has been largely ignored by the graphics community, even though it is a signature movement of the human body and an indicator for lifelike motion. In this paper, we present an anatomically inspired, physically based model of the human torso for the visual simulation of respiration using a mixed system of rigid and deformable parts. This novel composition of anatomical components is necessary to capture the key characteristics of breathing motion visible in the human trunk because the movement is generated fundamentally through the combination of both rigid bone and soft tissue. We propose a simple anatomically meaningful muscle element based on springs, which is used throughout both actively to drive the motion of the ribs and diaphragm and passively for other muscles like those of the abdomen. In addition, we introduce a straightforward method for preserving incompressible volume in deformable bodies to use in approximating the motion of the gut related to breath. Through the careful construction of this anatomically based torso, control for respiration becomes the generation of periodic contraction signals for a minimal set of two muscle groups. We show the flexibility of our approach through the animation of several breathing styles using our system and we verify our results through video and analytical comparisons.

Fast and Deep Deformation Approximations

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 characters 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 rigs 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.

State of animation tools in the industry

DreamWorks Animation, Pixar and Disney have invested in next-gen animation tools, or enhanced existing animation tools, for their animators. Autodesk has made a push in recent years to add and improve animation features in Maya. Rhythm and Hues built an extensible framework, Voodoo, that is used across multiple departments and received a 2013 Technical Achievement Academy Award.