Dataflow and Remapping for Wavelet Compression and Realtime View-Dependent Optimization of Billion-Triangle Isosurfaces

Currently, large physics simulations produce 3D fields whose individual surfaces, after conventional extraction processes, contain upwards of hundreds of millions of triangles. Detailed interactive viewing of these surfaces requires powerful compression to minimize storage, and fast view-dependent optimization of display triangulations to drive high-performance graphics hardware. In this work we provide an overview of an end-to-end multiresolution dataflow strategy whose goal is to increase efficiencies in practice by several orders of magnitude. Given recent advancements in subdivision-surface wavelet compression and view-dependent optimization, we present algorithms here that provide the “glue” that makes this strategy hold together. Shrink-wrapping converts highly detailed unstructured surfaces of arbitrary topology to the semi-structured form needed for wavelet compression. Remapping to triangle bintrees minimizes disturbing “pops” during realtime display-triangulation optimization and provides effective selectivetransmission compression for out-of-core and remote access to these huge surfaces.