cs.GRMay 25, 2026

Learning View-Dependent Splatting Kernels

Authors: Huakeng DingZhanpeng LiuFan PeiKun ZhouHongzhi Wu

Organizations: State Key Lab of CAD and CG, Zhejiang University, China · Hangzhou Research Institute of Holographic and AI Technology, China

Abstract

We present a differentiable framework to automatically learn view-dependent 2D kernels in a splatting-based pipeline to improve reconstruction quality and representation efficiency for novel 3D view synthesis. Our volumetric primitive is defined as a bounding ellipsoid and a 3D-kernel latent vector. We first learn a projection network to output a 2D-kernel latent, taking the attributes of the ellipsoid and the 3D-kernel latent as input. Next, the result is sent to a decoder to produce a radially symmetric 2D kernel in terms of Mahalanobis distance, bounded by the projected ellipsoid. The neural networks along with per-primitive attributes are jointly optimized. The effectiveness of our approach is demonstrated on standard benchmarks, comparing favorably against state-of-the-art techniques on both analytical and learned kernels. Finally, we extend the idea to learn general 2D kernels for 2D splatting as well as image representation.

Explore similar work

CardsList
  1. 3D Skew-Normal Splatting

    May 14, 2026Xiangru Wu, Ke Fan, Yanwei FuNovel View Synthesis3D Priors