MMGS: 10$\times$ Compressed 3DGS through Optimal Transport Aggregation based on Multi-view Ranking

TL;DR

The paper introduces MMGS, a novel 3D Gaussian compression framework that uses optimal transport and multi-view ranking to significantly reduce primitives while maintaining high-quality rendering.

Contribution

It presents a global geometric distribution matching approach with multi-view ranking and OT-based aggregation for efficient 3D Gaussian compression.

Findings

Achieves 10× faster training compared to vanilla 3DGS.

Reduces primitives to 10% while maintaining quality.

State-of-the-art rendering quality with compressed primitives.

Abstract

While 3D Gaussian Splatting (3DGS) has revolutionized 3D reconstruction, it suffers from significant overhead due to massive redundant primitives. Existing compression methods typically rely on local sampling or fixed pruning thresholds, which often struggle to balance redundancy reduction with high-fidelity rendering. To address this, we propose a novel framework that formulates Gaussian optimization as a global geometric distribution matching problem. Specifically, our approach integrates three components: (1) we introduce a multi-view 3D Gaussian contribution ranking mechanism that filters primitives using geometric consistency instead of local heuristics; (2) we propose a global Optimal Transport (OT)-based aggregation algorithm that merges redundant primitives while preserving the underlying geometry; and (3) we design an OT-based densification operator that maintains the Gaussian's distributional properties for stable optimization. Our approach achieves state-of-the-art rendering quality with only \textbf{10$\%$} primitives and \textbf{10$\times$} accelerated training speeds compared to vanilla 3DGS.