Donald H. House

Predicting the drape of woven cloth using interacting particles

We demonstrate a physically-based technique for predicting the drape of a wide variety of woven fabrics. The approach exploits a theoretical model that explicitly represents the microstructure of woven cloth with interacting particles, rather than utilizing a continuum approximation. By testing a cloth sample in a Kawabata fabric testing device, we obtain data that is used to tune the models energy functions, so that it reproduces the draping behavior of the original material. Photographs, comparing the drape of actual cloth with visualizations of simulation results, show that we are able to reliably model the unique large-scale draping characteristics of distinctly different fabric types.

Cloth modeling and animation

Written by leaders in the field of computer clothing design and simulation, Cloth Modeling and Animation is a vital resource for researchers and developers of cloth simulation software as well as computer animators and graphics programmers. Readers will learn about cloths nature and structure, scientific approaches to understanding its behavior and look, and the latest modeling and simulation techniques for automatically animating cloth on the computer.

Emotionally expressive agents

The ability to express emotions is important for creating believable interactive characters. To simulate emotional expressions in an interactive environment, an intelligent agent needs both an adaptive model for generating believable responses, and a visualization model for mapping emotions into facial expressions. Recent advances in intelligent agents and in facial modeling have produced effective algorithms for these tasks independently. We describe a method for integrating these algorithms to create an interactive simulation of an agent that produces appropriate facial expressions in a dynamic environment. Our approach to combining a model of emotions with a facial model represents a first step towards developing the technology of a truly believable interactive agent which has a wide range of applications from designing intelligent training systems to video games and animation tools.

A physically-based particle model of woven cloth

Every time a tablecloth is draped over a table it will fold and pleat in unique ways. We report on a physically-based model and a simulation methodology, which when used together are able to reproduce many of the attributes of this characteristic behavior of cloth. Our model utilizes a microscopic particle representation that directly treats the mechanical constraints between the threads in woven material rather than using a macroscopic continuum approximation. The simulation technique is hybrid, employing force methods for gross movement and energy methods to enforce constraints within the material. The model is developed and demonstrated within a visualization environment that allows full interaction between the simulated material and conventional constructive-solid-geometry models.

Wave particles

We present a new method for the real-time simulation of fluid surface waves and their interactions with floating objects. The method is based on the new concept of wave particles, which offers a simple, fast, and unconditionally stable approach to wave simulation. We show how graphics hardware can be used to convert wave particles to a height field surface, which is warped horizontally to account for local wave-induced flow. The method is appropriate for most fluid simulation situations that do not involve significant global flow. It is demonstrated to work well in constrained areas, including wave reflections off of boundaries, and in unconstrained areas, such as an ocean surface. Interactions with floating objects are easily integrated by including wave forces on the objects and wave generation due to object motion. Theoretical foundations and implementation details are provided, and experiments demonstrate that we achieve plausible realism. Timing studies show that the method is scalable to allow simulation of wave interaction with several hundreds of objects at real-time rates.

Mesh colors

The coloring of 3D models using 2D or 3D texture mapping has well-known intrinsic problems, such as mapping discontinuities and limitations to model editing after coloring. Workarounds for these problems often require adopting very complex approaches. Here we propose a new technique, called mesh colors, for associating color data directly with a polygonal mesh. The approach eliminates problems deriving from using a map from texture space to model space. Mesh colors is an extension of vertex colors where, in addition to keeping color values on each vertex, they are also kept on edges and faces. Like texture mapping, the approach allows higher texture resolution than model resolution, but at the same time it guarantees one-to-one correspondence between the model surface and the color data, and eliminates discontinuities. We show that mesh colors integrate well with the current graphics pipeline and can be used to generate very high-quality textures.

Better with bubbles: enhancing the visual realism of simulated fluid

We present a method for including the visual effect of bubbles in a computer graphics fluid simulation, thus enhancing the illusion of realism for a splashing fluid. Previous fluid simulation methods have not included bubbles. Bubble creation is integrated into the particle level-set fluid simulation algorithm. Individual bubbles are approximated by spheres, which form more complex shapes where they intersect. The rendering of bubbles and fluid are integrated to create the appearance of one continuous surface. At the fluid-air boundary, we integrate bubbles whenever level-set marker particles pass from from the outside to the inside of the fluid. Thus, these particles represent air that has become trapped within the fluid surface. In addition, we detect empty pockets within the fluid, that are often formed due to turbulence, and create bubbles within this space. This is an inexpensive way of giving the impression that the air trapped in air pockets has become bubbles. Photo-realistic images of simulation results are rendered with a raytracer that has been enhanced to include caustics, and to handle bubble-bubble interfaces. Comparison of these images with images rendered without bubbles supports our position that the simple addition of bubbles to a fluid simulation greatly enhances visual realism.

Continuous cartogram construction

Area cartograms are used for visualizing geographically distributed data by attaching measurements to regions of a map and scaling the regions such that their areas are proportional to the measured quantities. A continuous area cartogram is a cartogram that is constructed without changing the underlying map topology. We present a new algorithm for the construction of continuous area cartograms that was developed by viewing their construction as a constrained optimization problem. The algorithm uses a relaxation method that exploits hierarchical resolution, constrained dynamics, and a scheme that alternates goals of achieving correct region areas and adjusting region shapes. It is compared favorably to existing methods in its ability to preserve region shape recognition cues, while still achieving high accuracy.

Towards simulating cloth dynamics using interacting particles

Reviews a new approach being developed for modelling the dynamic behaviour of cloth. This work extends the cloth‐particle static draping model of Breen and House to include dynamics, and extends constrained dynamics simulation techniques developed by Witkin, Gleicher and Welch to yield performance enhancements. Fundamental to this approach is a new hierarchical approximation algorithm for constrained dynamics simulation which, it is hoped, will reduce the computational time demands of the algorithm to near real‐time range.

Adaptive particles for incompressible fluid simulation

We propose a particle-based technique for simulating incompressible fluid that includes adaptive refinement of particle sampling. Each particle represents a mass of fluid in its local region. Particles are split into several particles for finer sampling in regions of complex flow. In regions of smooth flow, neighboring particles can be merged. Depth below the surface and Reynolds number are exploited as our criteria for determining whether splitting or merging should take place. For the fluid dynamics calculations, we use the hybrid FLIP method, which is computationally simple and efficient. Since the fluid is incompressible, each particle has a volume proportional to its mass. A kernel function, whose effective range is based on this volume, is used for transferring and updating the particle’s physical properties such as mass and velocity. Our adaptive particle-based simulation is demonstrated in several scenarios that show its effectiveness in capturing fine detail of the flow, where needed, while efficiently sampling regions where less detail is required.