Confident Splatting: Confidence-Based Compression of 3D Gaussian Splatting via Learnable Beta Distributions

2025-06-28

AI Summary

To address the high storage and computational overhead caused by excessively large point clouds in 3D Gaussian Splatting rendering, this paper proposes a learnable Beta-distribution-based confidence modeling and compression framework. We introduce end-to-end differentiable, point-wise confidence scores, jointly optimized with a reconstruction-aware loss to enable adaptive pruning of low-confidence Gaussians. Unlike prior methods, our approach is agnostic to specific Gaussian lattice structures and thus compatible with diverse Splatting variants. Moreover, the scene-level average confidence serves as a novel, interpretable metric for reconstruction quality assessment. Extensive experiments demonstrate that our method achieves state-of-the-art visual fidelity while significantly reducing point cloud size—yielding superior compression-fidelity trade-offs compared to existing techniques. The source code and datasets are publicly available.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

3D Gaussian Splatting enables high-quality real-time rendering but often produces millions of splats, resulting in excessive storage and computational overhead. We propose a novel lossy compression method based on learnable confidence scores modeled as Beta distributions. Each splat's confidence is optimized through reconstruction-aware losses, enabling pruning of low-confidence splats while preserving visual fidelity. The proposed approach is architecture-agnostic and can be applied to any Gaussian Splatting variant. In addition, the average confidence values serve as a new metric to assess the quality of the scene. Extensive experiments demonstrate favorable trade-offs between compression and fidelity compared to prior work. Our code and data are publicly available at https://github.com/amirhossein-razlighi/Confident-Splatting

Problem

• Reducing storage and computational overhead in 3D Gaussian Splatting
• Optimizing confidence scores for splat pruning via Beta distributions
• Balancing compression and visual fidelity in 3D rendering

Innovation

• Confidence-based compression using learnable Beta distributions
• Reconstruction-aware loss optimizes splat confidence scores
• Architecture-agnostic pruning preserves visual fidelity