Abstract
Stochastic order-independent transparency enables efficient and elegant rendering of primitive-based radiance fields like 3D Gaussian Splatting models, but remains impractical due to the inherent visible noise in the output. We propose a principled approach to minimize high-frequency noise, addressing its sources at the representation and image synthesis level. During stochastic rendering, our history-based spatial resampling scheme drastically accelerates image convergence, while temporal importance resampling ensures coherence under camera movement. During training, a color regularizer implicitly reduces the variance along view rays in the 3DGS models. With these properties, our optimized, Vulkan-based renderer effectively mitigates output noise at low and high sample counts, achieving a substantial 13~dB PSNR increase in quality over previous stochastic methods at 1 sample per pixel and quickly converging to sorted 3DGS with an average L1 error of less than $10^{-4}$.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Windisch, F., Köhler, T., Radl, L., Wyman, C., Kopanas, G., Kerbl, B., & Steinberger, M. (2026). SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering. https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering
MLA 9
Windisch, Felix, et al. "SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering." https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering.
Chicago (author–date)
Windisch, Felix, Thomas Köhler, Lukas Radl, Chris Wyman, Georgios Kopanas, Bernhard Kerbl, and Markus Steinberger. 2026. "SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering." https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering.
Harvard
Windisch, F., Köhler, T., Radl, L., Wyman, C., Kopanas, G., Kerbl, B. and Steinberger, M. (2026) 'SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering', Available at: https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering.
Vancouver
Windisch F, Köhler T, Radl L, Wyman C, Kopanas G, Kerbl B, et al. SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering. https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering
IEEE
F. Windisch, T. Köhler, L. Radl, C. Wyman, G. Kopanas, B. Kerbl, and M. Steinberger, "SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering," https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering.