3D Gaussian Modeling and Ray Marching of OpenVDB datasets for Scientific Visualization

2025-09-14

AI Summary

To address the high memory overhead of dense grid data and the lack of a unified sparse modeling framework in scientific visualization, this paper proposes a novel 3D Gaussian particle modeling and rendering paradigm based on OpenVDB. Methodologically, it pioneers the use of OpenVDB/NanoVDB as a sparse foundation for scientific volumetric data, uniformly converting diverse data sources—including regular grids, adaptive mesh refinement (AMR) datasets, and point clouds—into compact 3D Gaussian representations. Rendering is performed via ray integration using OptiX 8.1, enhanced by hybrid distance-driven accumulation (HDDA) stepping and optical depth accumulation for improved efficiency. Contributions include: (1) the first OpenVDB-driven Gaussian modeling framework tailored for scientific visualization; (2) a unified sparse representation across heterogeneous data types with high compression ratios; and (3) rendering quality comparable to conventional methods, yet with significantly reduced memory consumption and accelerated rendering speed. Extensive experiments demonstrate effectiveness and generalizability across multiple real-world scientific datasets.

Technology Category

• Computer Vision: 3D Computer Vision
• Data Mining & Knowledge Management: Data Visualization & Summarization
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Data transparency and provenance
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

3D Gaussians are currently being heavily investigated for their scene modeling and compression abilities. In 3D volumes, their use is being explored for representing dense volumes as sparsely as possible. However, most of these methods begin with a memory inefficient data format. Specially in Scientific Visualization(SciVis), where most popular formats are dense-grid data structures that store every grid cell, irrespective of its contribution. OpenVDB library and data format were introduced for representing sparse volumetric data specifically for visual effects use cases such as clouds, fire, fluids etc. It avoids storing empty cells by masking them during storage. It presents an opportunity for use in SciVis, specifically as a modeling framework for conversion to 3D Gaussian particles for further compression and for a unified modeling approach for different scientific volume types. This compression head-start is non-trivial and this paper would like to present this with a rendering algorithm based on line integration implemented in OptiX8.1 for calculating 3D Gaussians contribution along a ray for optical-depth accumulation. For comparing the rendering results of our ray marching Gaussians renderer, we also implement a SciVis style primary-ray only NanoVDB HDDA based ray marcher for OpenVDB voxel grids. Finally, this paper also explores application of this Gaussian model to formats of volumes other than regular grids, such as AMR volumes and point clouds, using internal representation of OpenVDB grid class types for data hierarchy and subdivision structure.

Problem

• Compressing dense scientific volumetric data using sparse 3D Gaussian representations
• Developing ray marching rendering for OpenVDB datasets with optical-depth accumulation
• Extending Gaussian modeling to non-uniform volume types like AMR and point clouds

Innovation

• OpenVDB sparse volume conversion to 3D Gaussians
• OptiX-based ray marching with optical-depth accumulation
• Unified Gaussian modeling for multiple volume formats