Pixar

Ryan Kautzman

Jiayi Chong

William F Sheffler

Michael B Comet

Patrick Coleman

Michael K O'Brien

Gordon Cameron

Dave Hale

Geoffrey Irving

Jonathan Page

The disclosure provides an approach for a hybrid binding of meshes. Multiple meshes having levels of detail appropriate for different regions of a model are topologically connected by binding them together at simulation time. In one embodiment, a simulation application creates both geometric and force bindings between vertices in meshes. The simulation application identifies embedded vertices of a first mesh to be bound to a second mesh as being 'best' bound vertices, such as vertices coincident with vertices in the second mesh, and geometrically binds those vertices to appropriate vertices of the second mesh. The simulation application then binds each of the remaining embedded vertices which cannot be geometrically bound to vertices of the second mesh via a force binding, in which a zero-length spring force based technique is used to transfer forces and velocities between the force bound vertex of the first mesh and vertices of the second mesh.

Hybrid binding of meshes

A simulation application simulates a fluid flow. The simulation application divides a fluid into coarse volumetric elements (voxels) at a first resolution, and simulates the fluid flow at the first resolution. The simulation application defines a three-dimensional window that includes a portion of the coarse voxels. The simulation application divides each coarse voxel in the portion of the coarse voxels that lie within the three-dimensional window into a plurality of fine voxels at a second resolution. The simulation application simulates the fluid flow associated with each fine voxel in the plurality of fine voxels at the second resolution based at least in part on the simulation at the coarse resolution.

Windowed simulation in fluid flows

The movement of skin on an animated target, such as a character or other object, is simulated via a simulation software application. The software application creates a finite element model (FEM) comprising a plurality of finite elements based on an animated target. The software application attaches a first constraint force to a node associated with a first finite element in the plurality of finite elements. The software application attaches a second constraint force to the node. The software application detects a movement of the first finite element that results in a corresponding movement of the node. The software application determines a new position for the node based on the movement of at least one of the first finite element, the first constraint force, and the second constraint force.

Skin and flesh simulation using finite elements, biphasic materials, and rest state retargeting

This disclosure relates to computer-generated imagery (CGI) and computer-aided animation. More specifically, this disclosure relates to techniques for skin and volume simulation for use in CGI and computer-aided animation.

Systems and methods for generating skin and volume details for animated characters

Systems and methods for simulating ballistic motion on an animated object by continuously defining rest poses of the animation object in a motion simulator. Tetrahedral finite element simulation may be used with control mechanisms that target the simulation pose towards the animation. A simulation mesh is generated for two or more animated poses based on a first simulation mesh corresponding to a first pose of the animated object. The simulation meshes of the two or more animated poses are provided to a simulator for use by the simulator such that in the absence of external force and acceleration the simulator output approximates the animated poses. Embodiments of the present invention are particularly useful for fleshy, blobby animation objects such as human characters, although the techniques can be used for other objects having different characteristics.

Growing community of inventors

San Francisco, CA, United States of America

Ryan Kautzman