Point-Based and Image-Based Volumetric Rendering and Detail Modeling For Volume Graphics

Suny At Stony Brook, Stony Brook NY

Investigators

Abstract

"CAREER: Point-Based and Image-Based Volumetric Rendering and Detail Modeling For Volume Graphics" This project will advance and promote the emerging field of volume graphics for science, engineering, medicine, education, and entertainment. It will develop new data representations, rendering techniques, and modeling schemes that are all geared toward making the processing of volumetric datasets more feasible and realistic in a wide range of application domains. Volumetric data - essentially, stored values for every point in a volume - has been most popular in computational science and medicine. For example, a mathematical simulation in computational fluid dynamics produces pressure and speed of the fluid at every point. Similarly, medical scanners like MRI, PET, SPECT, and CT produce a value at every point in the body that they scan. Recently, the size of these datasets has grown at an alarming pace. Medical datasets of over one billion volume elements (voxels) have become frequent, and simulations will soon produce datasets that require terabytes to petabytes of storage. The need to develop efficient visualization methods for these has been identified as an immediate national interest. Technically, this project addresses two critical issues in the larger field of volume visualization. The first is the persistent lack of interactive volume rendering, which is hampered by the immense computational requirements of even moderate-sized datasets. The other is the inherently discrete nature of volumetric datasets. This is important because it establishes a boundary between reality and model, and pushing this boundary therefore provides increased realism in applications. More concretely, the work includes: Point-based representations of data, Point-based volumetric objects, Image-based rendering to assist in volume rendering and, Subdivision volumes with detail-on-demand. This research agenda is accompanied by a solid education plan that provides students with comprehensive knowledge in fields pertinent for visualization, volume rendering, and volume graphics. These fields are rather diverse, ranging from image processing to user interfaces to computer vision. Teamwork among the participating students is promoted by integrating all the research into a common application.