Research map / 3D vision / Gaussian splatting
MEGS^{2}: Memory-Efficient Gaussian Splatting via Spherical Gaussians and Unified Pruning
Jiarui Chen, Yikeng Chen, Yingshuang Zou, Ye Huang, Peng Wang, Yuan Liu, Yujing Sun, Wenping Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use soft (ADMM-style) pruning instead of hard pruning (ablation) | MEGS2 · Mip-NeRF360 | PSNR: 27.23 -> 27.54 | Improved |
| use soft pruning instead of hard pruning (ablation) | MEGS2 · Mip-NeRF360 | rendering VRAM (MB): 288 -> 265 | Improved |
| jointly optimize primitive and lobe pruning (unified) instead of sequential (ablation) | MEGS2 · Mip-NeRF360 | PSNR: 27.33 -> 27.54 | Improved |
| jointly optimize unified instead of sequential (ablation) | MEGS2 · Mip-NeRF360 | rendering VRAM (MB): 328 -> 265 | Improved |
| add color compensation after lobe removal (ablation) | MEGS2 · Mip-NeRF360 | PSNR: 27.46 -> 27.54 | Improved |
| combine GaussianSpa with Reduced3DGS SH pruning (ablation) | GaussianSpa · Mip-NeRF360 | PSNR: 27.54 -> 26.05 | Hurt |
| replace SH with SG in GaussianSpa (without unified pruning) (ablation) | GaussianSpa · Mip-NeRF360 | PSNR: 27.01 -> 27.54 | Improved |
| use arbitrarily oriented SG lobes instead of fixed orthogonal axes (ablation) | MEGS2 · Mip-NeRF360 | PSNR: about 0.6 dB drop for fixed-axis | Improved |
| replace SH with arbitrarily oriented SG color (ablation) | 3DGS · Mip-NeRF360 | quality-VRAM trade-off | Improved |
| MEGS2 vs GaussianSpa on rendering memory | MEGS2 HQ · Mip-NeRF360 | rendering VRAM (MB): 448 -> 265 | Improved |
| MEGS2 vs 3DGS on DeepBlending quality | MEGS2 HQ · DeepBlending | PSNR: 29.71 -> 30.17 | Improved |
| MEGS2 vs 3DGS on Tanks&Temples PSNR | MEGS2 HQ · Tanks&Temples | PSNR: 23.68 -> 23.45 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GaussianSpa — GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025) · better on 6 of 8 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.92)
- Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis (CVPR 2024 · similarity 0.91)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- Temporal Smoothness-Aware Rate-Distortion Optimized 4D Gaussian Splatting (NeurIPS 2025 · similarity 0.91)
- Fast Feedforward 3D Gaussian Splatting Compression (ICLR 2025 · similarity 0.91)
- Gaussian Herding across Pens: An Optimal Transport Perspective on Global Gaussian Reduction for 3DGS (NeurIPS 2025 · similarity 0.91)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
