[
    {
        "id": "osp-18034",
        "type": "article-journal",
        "title": "SteadySplats: Resampling of Low-Variance Gaussians for High-Fidelity Stochastic Rendering",
        "author": [
            {
                "family": "Windisch",
                "given": "Felix"
            },
            {
                "family": "Köhler",
                "given": "Thomas"
            },
            {
                "family": "Radl",
                "given": "Lukas"
            },
            {
                "family": "Wyman",
                "given": "Chris"
            },
            {
                "family": "Kopanas",
                "given": "Georgios"
            },
            {
                "family": "Kerbl",
                "given": "Bernhard"
            },
            {
                "family": "Steinberger",
                "given": "Markus"
            }
        ],
        "URL": "https://omanscience.com/en/articles/steadysplats-resampling-of-low-variance-gaussians-for-high-fidelity-stochastic-rendering",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "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 $10^{-4}$."
    }
]