Multivariate volume visualization through dynamic projections

摘要: We propose a multivariate volume visualization framework that tightly couples dynamic projections with high-dimensional transfer function design for interactive visualization. assume the complex, data in attribute space can be well-represented through collection of low-dimensional linear subspaces, and embed points variety 2D views created as onto these subspaces. Through projections, we present animated transitions between different to help user navigate explore effective design. Our not only provides more intuitive understanding but also allows under multiple views, which is flexible than being restricted single static view data. For large volumetric datasets, maintain interactivity during via intelligent sampling scalable clustering. Using examples combustion climate simulations, demonstrate how our used visualize interesting structures space.