频率自适应锐度正则化：提升3D高斯溅射泛化能力

摘要

尽管3D高斯溅射（3DGS）在多数配置中表现卓越，但在少样本场景下由于对稀疏观测的过拟合，其在新视角上的泛化能力不足。我们从机器学习视角重新审视3DGS的优化过程，将新视角合成定义为对未见过视角的泛化问题——这一方向尚未被充分探索。我们提出频率自适应锐度正则化（FASR），通过重构3DGS的训练目标，引导3DGS收敛至更具泛化能力的解。虽然锐度感知最小化（SAM）同样通过降低损失景观的锐度来提升分类模型的泛化能力，但由于任务差异，直接将其应用于3DGS会存在次优问题。具体而言，过度正则化会阻碍高频细节的重建，而降低正则化强度又会导致对锐度的惩罚不足。为此，我们通过反映图像的局部频率来设定正则化权重和估计局部锐度时的邻域半径。该方法能有效避免新视角下的漂浮伪影，并重建SAM容易过度平滑的精细细节。在多种配置的数据集上，我们的方法持续提升了各类基线的性能。代码将在https://bbangsik13.github.io/FASR 发布。

English

Despite 3D Gaussian Splatting (3DGS) excelling in most configurations, it lacks generalization across novel viewpoints in a few-shot scenario because it overfits to the sparse observations. We revisit 3DGS optimization from a machine learning perspective, framing novel view synthesis as a generalization problem to unseen viewpoints-an underexplored direction. We propose Frequency-Adaptive Sharpness Regularization (FASR), which reformulates the 3DGS training objective, thereby guiding 3DGS to converge toward a better generalization solution. Although Sharpness-Aware Minimization (SAM) similarly reduces the sharpness of the loss landscape to improve generalization of classification models, directly employing it to 3DGS is suboptimal due to the discrepancy between the tasks. Specifically, it hinders reconstructing high-frequency details due to excessive regularization, while reducing its strength leads to under-penalizing sharpness. To address this, we reflect the local frequency of images to set the regularization weight and the neighborhood radius when estimating the local sharpness. It prevents floater artifacts in novel viewpoints and reconstructs fine details that SAM tends to oversmooth. Across datasets with various configurations, our method consistently improves a wide range of baselines. Code will be available at https://bbangsik13.github.io/FASR.

频率自适应锐度正则化：提升3D高斯溅射泛化能力

Frequency-Adaptive Sharpness Regularization for Improving 3D Gaussian Splatting Generalization

摘要

Support