MMGS: 10$\times$ Compressed 3DGS through Optimal Transport Aggregation based on Multi-view Ranking
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916025606340608 |
|---|---|
| author | Zhao, Beizhen Yu, Sicheng Yin, Ziran Shen, Dongxu Wang, Hao |
| author_facet | Zhao, Beizhen Yu, Sicheng Yin, Ziran Shen, Dongxu Wang, Hao |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_19304 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MMGS: 10$\times$ Compressed 3DGS through Optimal Transport Aggregation based on Multi-view Ranking Zhao, Beizhen Yu, Sicheng Yin, Ziran Shen, Dongxu Wang, Hao Computer Vision and Pattern Recognition Graphics 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. |
| title | MMGS: 10$\times$ Compressed 3DGS through Optimal Transport Aggregation based on Multi-view Ranking |
| topic | Computer Vision and Pattern Recognition Graphics |
| url | https://arxiv.org/abs/2605.19304 |