Polynomial Voxel Maps for Time Varying Volumetric Rendering
Loading...
Links to Files
Permanent Link
Author/Creator
Author/Creator ORCID
Date
2018-01-01
Type of Work
Department
Computer Science and Electrical Engineering
Program
Computer Science
Citation of Original Publication
Rights
Distribution Rights granted to UMBC by the author.
Access limited to the UMBC community. Item may possibly be obtained via Interlibrary Loan thorugh a local library, pending author/copyright holder's permission.
This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
Access limited to the UMBC community. Item may possibly be obtained via Interlibrary Loan thorugh a local library, pending author/copyright holder's permission.
This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
Abstract
Volume visualization is used in a number of real-world applications, including scientific data representation, medical imaging, simulation display, and computer gaming. The standard method for rendering volumetric datasets in real time is Ray Marching. This is computationally expensive, and thus is almost exclusively offloaded to a discrete GPU. Modern graphics cards can handle the rendering costs, but the algorithm still does not account for time-varying datasets, in which the temporal changes in the data are relevant to the viewer. To handle pre-computed time-varying datasets, multiple snapshots of the data must be loaded to the GPU in real time, and interpolated Datasets with many time steps can be too large to fit in GPU memory at once, but too slow to load one step at a time. I present a method to reduce or eliminate runtime loading, instead preprocessing all of the data instances before passing them to the GPU. I do so with a volumetric analog to polynomial texture maps, by fitting a polynomial function of time to the data as it changes in each voxel. These functions are then evaluated on the GPU at runtime. This approach allows handling of large time-varying datasets, while barely affecting rendering performance. The result is the ability to visualize, interact with, and explore large temporal volume data, in real time.