MFA-DVR: Direct Volume Rendering of MFA Models
share
3D volume rendering is widely used to reveal insightful intrinsic patterns of volumetric datasets across many domains. However, the complex structures and varying scales of datasets make generating a high-quality volume rendering results efficiently a challenging task. Multivariate functional approximation (MFA) is a new data model that addresses some of the key challenges of volume visualization. MFA provides high-order evaluation of values and derivatives anywhere in the spatial domain, mitigating the artifacts caused by the zero- or first-order interpolation commonly implemented in existing volume visualization algorithms. MFA's compact representation improves the space complexity for large-scale volumetric data visualization, while its uniform representation of both structured and unstructured data allows the same ray casting algorithm to be used for a variety of input data types. In this paper, we present MFA-DVR, the first direct volume rendering pipeline utilizing the MFA model, for both structured and unstructured volumetric datasets. We demonstrate improved rendering quality using MFA-DVR on both synthetic and real datasets through a comparative study with raw and compressed data. We show that MFA-DVR not only generates more faithful volume rendering results with less memory footprint, but also performs faster than traditional algorithms when rendering unstructured datasets. MFA-DVR is implemented in the existing volume rendering pipeline of the Visualization Toolkit (VTK) in order to be accessible by the scientific visualization community.
READ FULL TEXT
