Search papers, labs, and topics across Lattice.
This paper introduces Spatially Varying Gaussian Splatting (SVGS), a novel approach to enhance Gaussian Splatting by incorporating spatially varying colors and opacity within each Gaussian primitive. SVGS explores bilinear interpolation, movable kernels, and tiny neural networks to model these spatially varying functions, enabling a more compact and expressive scene representation. Experiments demonstrate that SVGS, particularly with movable kernels, achieves superior novel view synthesis performance compared to the baseline, while preserving high-quality geometric reconstruction.
Gaussian Splatting gets a major upgrade: spatially-varying colors within each primitive dramatically boost novel view synthesis, especially for scenes with complex textures and simple geometry.
Gaussian Splatting demonstrates impressive results in multi-view reconstruction based on Gaussian explicit representations. However, the current Gaussian primitives only have a single view-dependent color and an opacity to represent the appearance and geometry of the scene, resulting in a non-compact representation. In this paper, we introduce a new method called SVGS (Spatially Varying Gaussian Splatting) that utilizes spatially varying colors and opacity in a single Gaussian primitive to improve its representation ability. We have implemented bilinear interpolation, movable kernels, and tiny neural networks as spatially varying functions. SVGS employs 2D Gaussian surfels as primitives, which significantly enhances novel-view synthesis while maintaining high-quality geometric reconstruction. This approach is particularly effective in practical applications, as scenes combining complex textures with relatively simple geometry occur frequently in real-world environments. Quantitative and qualitative experimental results demonstrate that all three functions outperform the baseline, with the best movable kernels achieving superior novel view synthesis performance on multiple datasets, highlighting the strong potential of spatially varying functions. Project page: https://ruixu.me/html/SuperGaussians/index.html
