# PUP 3D-GS: Principled Uncertainty Pruning for 3D Gaussian Splatting

arXiv: https://arxiv.org/abs/2406.10219 · HTML: https://mlsanity.com/paper/2406.10219/

## Key facts

- Cited by 95 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 35 references that are in the map.
- First posted on arXiv 14 Jun 2024.

## Builds on (most cited first, 24 of 24)

- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields](https://mlsanity.com/paper/2111.12077/) (2021)
- [NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections](https://mlsanity.com/paper/2008.02268/) (2020)
- [Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering](https://mlsanity.com/paper/2312.00109/) (2023)
- [LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS](https://mlsanity.com/paper/2311.17245/) (2023)
- [Compact 3D Gaussian Representation for Radiance Field](https://mlsanity.com/paper/2311.13681/) (2023)
- [PGSR: Planar-based Gaussian Splatting for Efficient and High-Fidelity Surface Reconstruction](https://mlsanity.com/paper/2406.06521/) (2024)
- [Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis](https://mlsanity.com/paper/2401.02436/) (2023)
- [Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussians](https://mlsanity.com/paper/2403.17898/) (2024)
- [Mini-Splatting: Representing Scenes with a Constrained Number of Gaussians](https://mlsanity.com/paper/2403.14166/) (2024)
- [HAC: Hash-grid Assisted Context for 3D Gaussian Splatting Compression](https://mlsanity.com/paper/2403.14530/) (2024)
- [EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingS](https://mlsanity.com/paper/2312.04564/) (2023)
- [Compact 3D Scene Representation via Self-Organizing Gaussian Grids](https://mlsanity.com/paper/2312.13299/) (2023)
- [ActiveNeRF: Learning where to See with Uncertainty Estimation](https://mlsanity.com/paper/2209.08546/) (2022)
- [RadSplat: Radiance Field-Informed Gaussian Splatting for Robust Real-Time Rendering with 900+ FPS](https://mlsanity.com/paper/2403.13806/) (2024)
- [RobustNeRF: Ignoring Distractors with Robust Losses](https://mlsanity.com/paper/2302.00833/) (2023)
- [Bayes' Rays: Uncertainty Quantification for Neural Radiance Fields](https://mlsanity.com/paper/2309.03185/) (2023)
- [Stochastic Neural Radiance Fields: Quantifying Uncertainty in Implicit 3D Representations](https://mlsanity.com/paper/2109.02123/) (2021)
- [CompGS: Efficient 3D Scene Representation via Compressed Gaussian Splatting](https://mlsanity.com/paper/2404.09458/) (2024)
- [Conditional-Flow NeRF: Accurate 3D Modelling with Reliable Uncertainty Quantification](https://mlsanity.com/paper/2203.10192/) (2022)
- [FisherRF: Active View Selection and Uncertainty Quantification for Radiance Fields using Fisher Information](https://mlsanity.com/paper/2311.17874/) (2023)
- [Density-aware NeRF Ensembles: Quantifying Predictive Uncertainty in Neural Radiance Fields](https://mlsanity.com/paper/2209.08718/) (2022)
- [NeU-NBV: Next Best View Planning Using Uncertainty Estimation in Image-Based Neural Rendering](https://mlsanity.com/paper/2303.01284/) (2023)
- [Unifying Approaches in Active Learning and Active Sampling via Fisher Information and Information-Theoretic Quantities](https://mlsanity.com/paper/2208.00549/) (2022)

## Built on by (most cited first, 30 of 95)

- [Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives](https://mlsanity.com/paper/2412.00578/) (2024)
- [FastGS: Training 3D Gaussian Splatting in 100 Seconds](https://mlsanity.com/paper/2511.04283/) (2025)
- [MaskGaussian: Adaptive 3D Gaussian Representation from Probabilistic Masks](https://mlsanity.com/paper/2412.20522/) (2024)
- [POp-GS: Next Best View in 3D-Gaussian Splatting with P-Optimality](https://mlsanity.com/paper/2503.07819/) (2025)
- [HDGS: Textured 2D Gaussian Splatting for Enhanced Scene Rendering](https://mlsanity.com/paper/2412.01823/) (2024)
- [Steepest Descent Density Control for Compact 3D Gaussian Splatting](https://mlsanity.com/paper/2505.05587/) (2025)
- [Modeling Uncertainty in 3D Gaussian Splatting through Continuous Semantic Splatting](https://mlsanity.com/paper/2411.02547/) (2024)
- [GoDe: Gaussians on Demand for Progressive Level of Detail and Scalable Compression](https://mlsanity.com/paper/2501.13558/) (2025)
- [EcoSplat: Efficiency-controllable Feed-forward 3D Gaussian Splatting from Multi-view Images](https://mlsanity.com/paper/2512.18692/) (2025)
- [Gaussian Herding across Pens: An Optimal Transport Perspective on Global Gaussian Reduction for 3DGS](https://mlsanity.com/paper/2506.09534/) (2025)
- [Improving 3D Gaussian Splatting Compression by Scene-Adaptive Lattice Vector Quantization](https://mlsanity.com/paper/2509.13482/) (2025)
- [SpeeDe3DGS: Speedy Deformable 3D Gaussian Splatting with Temporal Pruning and Motion Grouping](https://mlsanity.com/paper/2506.07917/) (2025)
- [RAP: Fast Feedforward Rendering-Free Attribute-Guided Primitive Importance Score Prediction for Efficient 3D Gaussian Splatting Processing](https://mlsanity.com/paper/2602.19753/) (2026)
- [Predictive Photometric Uncertainty in Gaussian Splatting for Novel View Synthesis](https://mlsanity.com/paper/2603.22786/) (2026)
- [ContraGS: Codebook-Condensed and Trainable Gaussian Splatting for Fast, Memory-Efficient Reconstruction](https://mlsanity.com/paper/2509.03775/) (2025)
- [Auto3R: Automated 3D Reconstruction and Scanning via Data-driven Uncertainty Quantification](https://mlsanity.com/paper/2512.04528/) (2025)
- [Prune Wisely, Reconstruct Sharply: Compact 3D Gaussian Splatting via Adaptive Pruning and Difference-of-Gaussian Primitives](https://mlsanity.com/paper/2602.24136/) (2026)
- [LEGO-SLAM: Language-Embedded Gaussian Optimization SLAM](https://mlsanity.com/paper/2511.16144/) (2025)
- [POTR: Post-Training 3DGS Compression](https://mlsanity.com/paper/2601.14821/) (2026)
- [GaussianPOP: Principled Simplification Framework for Compact 3D Gaussian Splatting via Error Quantification](https://mlsanity.com/paper/2602.06830/) (2026)
- [VarSplat: Uncertainty-aware 3D Gaussian Splatting for Robust RGB-D SLAM](https://mlsanity.com/paper/2603.09673/) (2026)
- [GeoNVS: Geometry Grounded Video Diffusion for Novel View Synthesis](https://mlsanity.com/paper/2603.14965/) (2026)
- [REFINE: Super-efficient 3D Gaussian Splatting Pruning via Rendering-Free Primitive Importance](https://mlsanity.com/paper/2606.09074/) (2026)
- [Trick-GS: A Balanced Bag of Tricks for Efficient Gaussian Splatting](https://mlsanity.com/paper/2501.14534/) (2025)
- [ControlGS: Consistent Structural Compression Control for Deployment-Aware Gaussian Splatting](https://mlsanity.com/paper/2505.10473/) (2025)
- [SVR-GS: Spatially Variant Regularization for Probabilistic Masks in 3D Gaussian Splatting](https://mlsanity.com/paper/2509.11116/) (2025)
- [Splatwizard: A Benchmark Toolkit for 3D Gaussian Splatting Compression](https://mlsanity.com/paper/2512.24742/) (2025)
- [SA-ResGS: Self-Augmented Residual 3D Gaussian Splatting for Next Best View Selection](https://mlsanity.com/paper/2601.03024/) (2026)
- [FreeFix: Boosting 3D Gaussian Splatting via Fine-Tuning-Free Diffusion Models](https://mlsanity.com/paper/2601.20857/) (2026)
- [NanoGS: Training-Free Gaussian Splat Simplification](https://mlsanity.com/paper/2603.16103/) (2026)

## Abstract (arXiv)

Recent advances in novel view synthesis have enabled real-time rendering speeds with high reconstruction accuracy. 3D Gaussian Splatting (3D-GS), a foundational point-based parametric 3D scene representation, models scenes as large sets of 3D Gaussians. However, complex scenes can consist of millions of Gaussians, resulting in high storage and memory requirements that limit the viability of 3D-GS on devices with limited resources. Current techniques for compressing these pretrained models by pruning Gaussians rely on combining heuristics to determine which Gaussians to remove. At high compression ratios, these pruned scenes suffer from heavy degradation of visual fidelity and loss of foreground details. In this paper, we propose a principled sensitivity pruning score that preserves visual fidelity and foreground details at significantly higher compression ratios than existing approaches. It is computed as a second-order approximation of the reconstruction error on the training views with respect to the spatial parameters of each Gaussian. Additionally, we propose a multi-round prune-refine pipeline that can be applied to any pretrained 3D-GS model without changing its training pipeline. After pruning 90% of Gaussians, a substantially higher percentage than previous methods, our PUP 3D-GS pipeline increases average rendering speed by 3.56× while retaining more salient foreground information and achieving higher image quality metrics than existing techniques on scenes from the Mip-NeRF 360, Tanks & Temples, and Deep Blending datasets.