ECoNGS: Efficient Compressive Neural Gaussian Splats for Volume Visualization

2026-07-20

AI Summary

Existing primitive-based volumetric visualization methods suffer from representation redundancy, large model sizes, and the need for per-scene optimization. This work proposes an efficiently compressed neural Gaussian splatting framework that leverages a lightweight neural network to dynamically predict implicit, editable Gaussians from explicit anchor points. The approach introduces cross-scene parameter sharing and neural entropy coding to compress anchor attributes, along with a volume-visualization-oriented Gaussian initialization strategy. By integrating the rendering efficiency of explicit representations with the compactness of implicit ones, the method significantly outperforms iVR-GS across diverse volumetric datasets—achieving up to a 2.2 dB improvement in PSNR, a 6.1× reduction in model size, and up to a 5.9× decrease in training time.

Abstract

Recent advances in differentiable Gaussian splatting have highlighted the potential of primitive-based approaches as alternative scene representations for interactive, high-quality, volume visualization (VolVis) of large datasets. However, the explicit nature of current primitive-based methods, combined with isolated optimization for each VolVis scene, results in redundant, non-compact representations. We present ECoNGS, an efficient compressive neural Gaussian splatting framework for VolVis scene representation. ECoNGS employs lightweight neural networks to dynamically predict implicit, editable Gaussian splats from explicit anchor points, effectively combining model compactness and parameter efficiency of implicit representations with high-performance rendering of explicit primitives. We explore a joint learning strategy that clusters geometrically similar scenes and shares parameters across them, significantly reducing overall training time and model size while maintaining reconstruction fidelity. To achieve a more compact scene representation, we further compress the explicit anchor attributes using a neural entropy model that estimates their probability distributions, enabling compact storage via entropy coding. We systematically investigate Gaussian initialization strategies and propose a simple yet effective scheme tailored for VolVis scenes, improving reconstruction accuracy and accelerating convergence. We evaluate ECoNGS qualitatively and quantitatively across various univariate and multivariate VolVis scenes, highlighting its superior performance over prior methods in training time, reconstruction quality, and model size. In particular, compared with the prior method iVR-GS, ECoNGS improves reconstruction quality by up to 2.2 dB in PSNR while reducing the model size by up to 6.1x and the training time by up to 5.9x. The code is available at https://github.com/TouKaienn/ECoNGS.

Problem

• volume visualization
• Gaussian splatting
• compact representation
• scene redundancy
• primitive-based methods

Innovation

• Neural Gaussian Splatting
• Volume Visualization
• Implicit-Explicit Hybrid Representation
• Neural Entropy Coding
• Joint Scene Learning