Hierarchical splatting of scattered 4D data

Abstract

To let scientists interactively view these data sets at high resolution on desktop workstations or PCs, we want to visualize the scattered data directly without resampling them to a volume density. We propose accelerating the visualization of scattered point data with a hierarchical data structure based on a principal component analysis (PCA) clustering procedure. By traversing this structure from each frame, we can trade off rendering speed verses image quality and use lower hierarchy levels during interaction. Our scheme also lets us interpolate the given point positions using cubic splines.