Doppler Volume Rendering: A Dynamic, Piecewise Linear Spectral Representation for Visualizing Astrophysics Simulations

Reem Alghamdi, Thomas Müller, Alberto Jaspe-Villanueva, Markus Hadwiger, and Filip Sadlo

Computer Graphics Forum Volume 42 (2023), Number 3
Presented at EuroVIS 2023

Paper PDF

@article{Alghamdi:2023:DVR},
 author = {Alghamdi, Reem and Müller, Thomas and Jaspe-Villanueva, Alberto and Hadwiger, Markus and Sadlo, Filip},
 title = {Doppler Volume Rendering: A Dynamic, Piecewise Linear Spectral Representation for Visualizing Astrophysics Simulations},
 journal = {Computer Graphics Forum},
 volume = {42},
 number = {3},
 pages = {39-49},
 keywords = {CCS Concepts, • Human-centered computing → Scientific visualization, • Computing methodologies → Simulation evaluation},
 doi = {https://doi.org/10.1111/cgf.14810},
 url = {https://onlinelibrary.wiley.com/doi/abs/10.1111/cgf.14810},
 eprint = {https://onlinelibrary.wiley.com/doi/pdf/10.1111/cgf.14810},
 year = {2023}
 }

Abstract

We present a novel approach for rendering volumetric data including the Doppler effect of light, which is crucial for the case of very high velocities of the matter that is emitting light. Similar to the acoustic Doppler effect, which is caused by the relative movement between a sound emitter and an observer, light waves as electromagnetic radiation also experience a compression or expansion when the light emitter and an observer move relative to one another. To take this effect into account, we employ spectral volume rendering in an emission/absorption model, with voxels emitting and attenuating light at different wavelengths. Our approach enables accurate computation of the Doppler effect of light in realtime volume rendering, with relative motion provided by an additional 3D velocity vector field. In particular, we propose a new representation for light spectra that uses a dynamic set of piecewise linear basis functions, tailored to accurate and efficient evaluation of the Doppler effect in volume rendering. Our basis functions are used during light accumulation along viewing rays to achieve higher efficiency and accuracy compared to point-based spectral representations. Our method is particularly useful for analyzing astrophysical simulations.