Describes an effective combination of visualization methods, such as superquadric tensor glyphs and direct volume rendering, and a mathematical framework for analysis of the gradient of the diffusion tensor field, to enable a generalized form of edge detection.

Proposes an intuitive tensor glyph based on superquadric surfaces, and demonstrates the glyph for visualizing diffusion tensor data of the brain.

Describes convolution-based curvature measurement in regularly sampled scalar fields, and shows how curvature information improves non-photorealistic volume rendering, and facilitates visualization of level set solutions and isosurface uncertainty.

Presents a simple method for matching the luminance of a color and shade of gray, and shows how this method can be employed to generate colormaps with a predetermined pattern of luminance variation, such as monotically increasing luminance, or isoluminance.

Demonstrates the effectiveness and utility of multi-dimensional transfer functions with a new set of interaction techniques and widgets (for making the specification of multi-dimensional transfer functions convenient and intuitive), and modern graphics hardware (for making volume rendering interactive).

Extended version of the Vis99 paper below. Describes more general transfer functions, blending of lit-tensors with surface shading, and the use of Alan Turing's reaction-diffusion textures as a tool for 2D and 3D tensor visualization. Also addresses issues of diffusion tensor data interpolation.

Explores the use of direct volume rendering for visualizating three-dimensional second-order diffusion tensor data, with ideas on how to color and illuminate diffusion tensors, as well as how to assign opacity to them.

A compact, 8-page version of my Master's thesis (below).

Demonstrates that for a common scalar field direct volume rendering task (visualization of boundaries between homogenous regions), the parameter space of transfer functions can be greatly simplified by analyzing a histogram of the data value and its two spatial derivatives.