用小波变换抑制高频噪声，让稀疏视角3D高斯点云重建更真实。
DWTGS: Rethinking Frequency Regularization for Sparse-view 3D Gaussian Splatting
- 改用小波域损失，只监督多尺度低频信息，避免过拟合高频细节。
- 在多个小波层级上仅保留低频子带，自监督约束高频子带稀疏性。
- 相比傅里叶方法，显著减少高频幻觉，提升新视角泛化能力。
稀疏视角3D高斯点云渲染（3DGS）在重建高质量新视角时面临挑战，常因过度拟合训练视图的高频细节而失真。现有频率正则化依赖傅里叶变换，参数难调且易诱发有害的高频学习。本文提出DWTGS框架，通过小波空间损失重新思考频率正则化：在多级离散小波变换（DWT）中，仅监督低频（LF）LL子带，同时以自监督方式强制高频（HF）HH子带稀疏。在多个基准测试中，DWTGS持续优于基于傅里叶的方法，该低频主导策略有效提升泛化性并减少高频幻觉。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。