Visual Fidelity of 3d Regular Sampling and Reconstruction

For generic 3D sampling, the Body Centered Cubic (BCC) lattice and the Face Centered Cubic (FCC) lattice are significantly superior to the traditionally popular Cartesian Cubic (CC) lattice. Motivated by the goal of high visual fidelity in the visualization community, this thesis investigates the relative perceptual merits of high quality reconstruction filters for CC, BCC, and FCC sampled data. We recruited 24 participants and gave them pairs of images to discriminate. For each pair, a participant is asked to choose the better of the two images: one image rendered from data sampled on the CC lattice, and one image rendered from data sampled on the BCC or FCC lattice. We used both synthetic and Computed Tomography (CT) volumetric data, and report that by using approximately 30 – 35% fewer samples, BCC and FCC reconstructed data result in images that exhibit comparable visual quality to their CC counterparts.