Multiresolution Visualization Tools
for the Interactive Exploration and Analysis
of Large-Scale Multidimensional
Experimental and Simulated Data
The goal of this project is to develop a computational pipeline which will facilitate the analysis and visualization of large volume data sets. Currently, scanning devices allow researchers to acquire enormous amounts of volume data on biological specimens and fluid-dynamic phenomena. This data is acquired over time, allowing researchers to examine dynamic processes and development, and producing sequences of volume datasets. The data stored at each voxel of the datasets may be scalar, vector, or tensor values. The project will develop the methods which will segment regions of interest in these datasets, and generate triangles meshes which define the surfaces of the segmented regions. The meshes will be compressed so that they may be transferred over networks and viewed on conventional workstations. Algorithms will be developed to directly render the multi-dimensional volume datasets. Novel display and interaction techniques will then be explored using emerging display systems. Developing and integrating these computational methods will significantly shorten the time needed to process and analyze the large amounts of data produced by today's scanning devices, accelerating the advances made in neuroscience and fluid-dynamics.