Files and links (1)
CC BY V4.0, Open Access
Preprint
SVR-GS: Spatially Variant Regularization for Probabilistic Masks in 3D Gaussian Splatting
ArXiv.org
Cornell University
2025
Abstract
3D Gaussian Splatting (3DGS) enables fast, high-quality novel view synthesis but typically relies on densification followed by pruning to optimize the number of Gaussians. Existing mask-based pruning, such as MaskGS, regularizes the global mean of the mask, which is misaligned with the local per-pixel (per-ray) reconstruction loss that determines image quality along individual camera rays. This paper introduces SVR-GS, a spatially variant regularizer that renders a per-pixel spatial mask from each Gaussian's effective contribution along the ray, thereby applying sparsity pressure where it matters: on low-importance Gaussians. We explore three spatial-mask aggregation strategies, implement them in CUDA, and conduct a gradient analysis to motivate our final design. Extensive experiments on Tanks\&Temples, Deep Blending, and Mip-NeRF360 datasets demonstrate that, on average across the three datasets, the proposed SVR-GS reduces the number of Gaussians by 1.79\(\times\) compared to MaskGS and 5.63\(\times\) compared to 3DGS, while incurring only 0.50 dB and 0.40 dB PSNR drops, respectively. These gains translate into significantly smaller, faster, and more memory-efficient models, making them well-suited for real-time applications such as robotics, AR/VR, and mobile perception.
Details
- Title
- SVR-GS: Spatially Variant Regularization for Probabilistic Masks in 3D Gaussian Splatting
- Authors/Creators
- Ashkan TaghipourVahid NaghshinBenjamin SouthwellFarid BoussaidHamid LagaMohammed Bennamoun
- Publication Details
- ArXiv.org
- Publisher
- Cornell University
- Identifiers
- 991005813047607891
- Copyright
- © the Authors
- Murdoch Affiliation
- Centre for Biosecurity and One Health; School of Information Technology; Centre for Healthy Ageing
- Language
- English
- Resource Type
- Preprint
Metrics
8 Record Views