ESVR: 3D Ellipsoid-based Sparse Volume Rendering via Structure-aware Primitive Learning and Per-primitive Ray Sampling

2026-08-05

AI Summary

This work addresses the challenge of efficiently representing and rendering large-scale sparse volumetric data, where existing approaches either incur substantial computational overhead or suffer from information loss and limited control over transfer functions due to reliance on image-space supervision. To overcome these limitations, we propose the first structure-aware ellipsoidal primitive rendering framework that learns directly from raw volumetric data. Our method introduces a differentiable, compactly supported ellipsoidal representation, a complementary pruning mechanism, and a per-primitive ray sampling strategy, further enhanced by chunk-wise optimization and “ghost ellipsoid” techniques to preserve boundary context. Evaluated on multiple large sparse datasets, the approach achieves compression ratios up to four orders of magnitude while maintaining competitive reconstruction quality and enabling real-time interactive rendering at 43–223 FPS.

Abstract

Efficient representation and rendering of large-scale sparse volumetric data remain challenging in scientific visualization, as meaningful structures often occupy only a small fraction of the spatial domain. While direct volume rendering (DVR) provides high-quality visualization, its computational and memory costs scale poorly with data size. Recent advances in 3D Gaussian Splatting (3DGS) address this challenge by representing volumetric scenes with compact geometric primitives, enabling efficient, high-fidelity rendering. However, existing 3DGS-based methods learn from DVR rendered images rather than raw volumes, leading to information loss and limiting flexible transfer function control for interactive exploration. To address these limitations, we propose ESVR, an ellipsoid-based sparse volume rendering framework that directly learns and renders volumetric data in 3D space. Our method combines differentiable ellipsoidal primitives with bounded support, structure-aware primitive learning with complementary pruning, and a per-primitive ray sampling strategy for fast and accurate transfer function mapping. To support large-scale datasets, we further introduce a chunk-based optimization scheme with ghost ellipsoids, providing boundary context during training. Across large sparse datasets, ESVR achieves up to four orders of magnitude compression and real-time rendering at 43-223 FPS while maintaining competitive reconstruction quality.

Problem

• sparse volume rendering
• 3D Gaussian Splatting
• transfer function
• large-scale volumetric data
• information loss

Innovation

• ellipsoid-based rendering
• structure-aware primitive learning
• per-primitive ray sampling
• sparse volume rendering
• differentiable primitives