4D Neural Voxel Splatting: Dynamic Scene Rendering with Voxelized Guassian Splatting

TL;DR

The paper introduces 4D Neural Voxel Splatting, a method combining voxel-based representations with neural Gaussian splatting to efficiently model and render dynamic scenes with reduced memory and faster training.

Contribution

It proposes a novel 4D neural voxel splatting approach that reduces memory overhead and accelerates training for dynamic scene rendering, outperforming existing methods.

Findings

Significant memory reduction compared to prior methods

Faster training times while maintaining high image quality

Enables real-time rendering with improved visual fidelity

Abstract

Although 3D Gaussian Splatting (3D-GS) achieves efficient rendering for novel view synthesis, extending it to dynamic scenes still results in substantial memory overhead from replicating Gaussians across frames. To address this challenge, we propose 4D Neural Voxel Splatting (4D-NVS), which combines voxel-based representations with neural Gaussian splatting for efficient dynamic scene modeling. Instead of generating separate Gaussian sets per timestamp, our method employs a compact set of neural voxels with learned deformation fields to model temporal dynamics. The design greatly reduces memory consumption and accelerates training while preserving high image quality. We further introduce a novel view refinement stage that selectively improves challenging viewpoints through targeted optimization, maintaining global efficiency while enhancing rendering quality for difficult viewing angles. Experiments demonstrate that our method outperforms state-of-the-art approaches with significant memory reduction and faster training, enabling real-time rendering with superior visual fidelity.