| |
|
|
|
Environment Visualization
| This is the
kind of thing I worked on before I fell into NewTek's
gravity well. Unlike a lot of 3D artists who start out with a desire to remake Star
Wars, I got into 3D to visualize smoke and dust plumes, and I created all of my
models by writing programs. If you're a 3D artist who's never heard of Gaussian puffs,
just think of them as metaballs on steroids. The visualization process begins with an atmospheric transport model, a computer program that simulates the movement of a cloud, or plume, through the air. Plumes are typically modeled as a set of Gaussian puffs, each of which is defined by a mass (how much stuff is in it), a centroid (where it is), and some number of spread moments (how fast the density falls off at increasing distances from the centroid). The falloff function is similar to one side of a normal distribution (bell curve), and the puffs themselves have an ellipsoidal shape. I resample this data onto a regular 3D grid to create a concentration field (a 3D array of density values). The density at each grid point is the sum of the contributions of each puff at that point. The contribution of a puff is calculated by finding the distance from the point to the centroid and then plugging that into the falloff function. (More or less. The "falloff" is formally defined using tensors, so I'm glossing over the math.) I then use a surface-building method that creates a "skin" around grid points with densities higher than a selected value. The resulting sheet is called an isolevel surface because all points on it correspond to the selected density. (Isolevel methods are discussed in the references at the end of this page. The one I use was developed to create 3D images from CT and MRI scans.) Once I have the geometry of a time sequence of surfaces, I can load it into LightWave and, for example, combine it with terrain data. I've applied this visualization approach as a consultant working with U.S. defense contractors and government agencies. These include Logicon/RDA, SAIC, and the Defense Threat Reduction Agency (formerly the Defense Special Weapons Agency, and before that, the Defense Nuclear Agency). I developed the computer programs under a Small Business Innovation Research (SBIR) contract for which I was the principal investigator. Information about the SBIR program is available from a number of U.S. government sites. Try SBA, NSF, and DARPA.
Foley, J., A. van Dam, S. Feiner, J. Hughes, Computer Graphics: Principles and Practice, 2nd ed. in C, Addison Wesley, 1996. |
 |