Research map / 3D vision / Gaussian splatting
SkipGS: Post-Densification Backward Skipping for Efficient 3DGS Training
Jingxing Li, Yongjae Lee, Deliang Fan
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 |
|---|---|---|---|
| add SkipGS backward gating | Vanilla 3DGS · Mip-NeRF 360 (avg) | total training time (s): 1705.7 -> 1311.1 | Improved |
| add SkipGS backward gating | Vanilla 3DGS · Mip-NeRF 360 (avg) | PSNR: 27.52 -> 27.52 | No clear effect |
| add SkipGS on top of Taming 3DGS | Taming 3DGS · Mip-NeRF 360 (avg) | post-densification time (s): 757.0 -> 392.0 | Improved |
| add SkipGS on top of FastGS | FastGS · Mip-NeRF 360 (avg) | total training time (s): 181.9 -> 164.5 | Improved |
| add SkipGS on top of GaussianSpa | GaussianSpa · Mip-NeRF 360 (avg) | total training time (s): 2640.9 -> 2490.9 | Improved |
| add SkipGS on top of Taming 3DGS | Taming 3DGS · Tanks and Temples (avg) | PSNR: 24.43 -> 24.34 | Hurt |
| add SkipGS to vanilla 3DGS | Vanilla 3DGS · Deep Blending (avg) | PSNR: 29.79 -> 29.88 | Improved |
| add backward budget control to SkipGS (ablation) | SkipGS on Taming 3DGS · Mip-NeRF 360 (avg) ablation | PSNR: 27.27 -> 27.92 | Improved |
| add backward budget control to SkipGS (ablation) | SkipGS on GaussianSpa · Mip-NeRF 360 (avg) ablation | PSNR: 27.23 -> 27.60 | Improved |
| add backward budget control to SkipGS (ablation) | SkipGS on Taming 3DGS · Mip-NeRF 360 (avg) ablation | post-densification time (s): 161.0 -> 392.0 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- GaussianSpa — GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025) · better on 3 of 3 shared results
- Speedy-Splat — Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025) · better on 3 of 3 shared results
- LightGaussian — LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS (NeurIPS 2024) · better on 3 of 3 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
- DropoutGS: Dropping Out Gaussians for Better Sparse-view Rendering (CVPR 2025 · similarity 0.91)
- DropGaussian: Structural Regularization for Sparse-view Gaussian Splatting (CVPR 2025 · similarity 0.91)
- 3DGS-Enhancer: Enhancing Unbounded 3D Gaussian Splatting with View-consistent 2D Diffusion Priors (NeurIPS 2024 · similarity 0.91)
- LP-3DGS: Learning to Prune 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.90)
- Quantifying and Alleviating Co-Adaptation in Sparse-View 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.90)
- GaussianSpa: An "Optimizing-Sparsifying" Simplification Framework for Compact and High-Quality 3D Gaussian Splatting (CVPR 2025 · similarity 0.89)
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.
