DWTGS: Rethinking Frequency Regularization for Sparse-view 3D Gaussian Splatting

TL;DR

DWTGS introduces a wavelet-based frequency regularization framework for sparse-view 3D Gaussian Splatting, improving view synthesis quality by better handling high-frequency details and reducing overfitting.

Contribution

It rethinks frequency regularization by using wavelet-space losses, focusing on low-frequency supervision and sparse high-frequency enforcement, outperforming Fourier-based methods.

Findings

DWTGS outperforms Fourier-based regularization methods across benchmarks.

Wavelet-space losses improve generalization and reduce high-frequency hallucinations.

LF-centric strategy enhances the quality of 3D Gaussian Splatting reconstructions.

Abstract

Sparse-view 3D Gaussian Splatting (3DGS) presents significant challenges in reconstructing high-quality novel views, as it often overfits to the widely-varying high-frequency (HF) details of the sparse training views. While frequency regularization can be a promising approach, its typical reliance on Fourier transforms causes difficult parameter tuning and biases towards detrimental HF learning. We propose DWTGS, a framework that rethinks frequency regularization by leveraging wavelet-space losses that provide additional spatial supervision. Specifically, we supervise only the low-frequency (LF) LL subbands at multiple DWT levels, while enforcing sparsity on the HF HH subband in a self-supervised manner. Experiments across benchmarks show that DWTGS consistently outperforms Fourier-based counterparts, as this LF-centric strategy improves generalization and reduces HF hallucinations.