Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction

TL;DR

Neural Harmonic Textures enhance primitive-based 3D reconstruction by applying harmonic analysis to improve detail modeling, achieving state-of-the-art real-time novel view synthesis with reduced computation.

Contribution

The paper introduces Neural Harmonic Textures, a novel harmonic analysis approach that improves detail modeling in primitive-based neural reconstruction.

Findings

Achieves state-of-the-art real-time novel view synthesis.

Reduces computational cost with a single neural network pass.

Seamlessly integrates into existing primitive-based pipelines.

Abstract

Primitive-based methods such as 3D Gaussian Splatting have recently become the state-of-the-art for novel-view synthesis and related reconstruction tasks. Compared to neural fields, these representations are more flexible, adaptive, and scale better to large scenes. However, the limited expressivity of individual primitives makes modeling high-frequency detail challenging. We introduce Neural Harmonic Textures, a neural representation approach that anchors latent feature vectors on a virtual scaffold surrounding each primitive. These features are interpolated within the primitive at ray intersection points. Inspired by Fourier analysis, we apply periodic activations to the interpolated features, turning alpha blending into a weighted sum of harmonic components. The resulting signal is then decoded in a single deferred pass using a small neural network, significantly reducing computational cost. Neural Harmonic Textures yield state-of-the-art results in real-time novel view synthesis while bridging the gap between primitive- and neural-field-based reconstruction. Our method integrates seamlessly into existing primitive-based pipelines such as 3DGUT, Triangle Splatting, and 2DGS. We further demonstrate its generality with applications to 2D image fitting and semantic reconstruction.