Research map / 3D vision / Gaussian splatting
SAGS: Structure-Aware 3D Gaussian Splatting
Evangelos Ververas, Rolandos Alexandros Potamias, Jifei Song, Jiankang Deng, Stefanos Zafeiriou
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 |
|---|---|---|---|
| add curvature-aware densification (ablation) | SAGS · Deep Blending | PSNR: 29.87 -> 30.47 | Improved |
| add GNN aggregation instead of MLP (ablation) | SAGS · Deep Blending | PSNR: 29.94 -> 30.47 | Improved |
| add GNN aggregation instead of MLP (ablation) | SAGS · Tanks&Temples | PSNR: 24.19 -> 24.88 | Improved |
| add curvature-aware densification (ablation) | SAGS · Tanks&Temples | PSNR: 23.97 -> 24.88 | Improved |
| add positional encoding of point positions (ablation) | SAGS · Deep Blending | PSNR: 30.21 -> 30.47 | Improved |
| add global feature g (ablation) | SAGS · Tanks&Temples | PSNR: 24.42 -> 24.88 | Improved |
| add view-dependent positions to appearance attributes (ablation) | SAGS · Deep Blending | PSNR: 30.07 -> 30.47 | Improved |
| structure-aware GNN model vs structure-agnostic Scaffold-GS | SAGS · Mip-NeRF360 | PSNR: 28.84 -> 29.65 | Improved |
| use lightweight midpoint-interpolation model (SAGS-Lite) instead of full SAGS | SAGS · Deep Blending | model size (MB): 58 -> 28 | Improved |
| use SAGS-Lite instead of full SAGS | SAGS · Deep Blending | PSNR: 30.47 -> 29.07 | Hurt |
| SAGS vs Scaffold-GS storage | SAGS · Mip-NeRF360 | model size (MB): 252 -> 135 | Improved |
Later papers that compared with this method
2 later accepted papers report SAGS as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Scene Experts: Specializing in 3D Gaussian Splatting with Adaptive Decomposition (AAAI 2026 · better on 4 of 4 shared results)
- CoRoGS: Contextual Gaussian Splatting for Robust Large-Deviation View Synthesis (CVPR 2026 · better on 2 of 2 shared results)
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.
- Scaffold-GS — Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering (CVPR 2024) · better on 4 of 4 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
- GaussianPro: 3D Gaussian Splatting with Progressive Propagation (ICML 2024 · similarity 0.94)
- Speedy-Splat: Fast 3D Gaussian Splatting with Sparse Pixels and Sparse Primitives (CVPR 2025 · similarity 0.93)
- Optimized Minimal 3D Gaussian Splatting (NeurIPS 2025 · similarity 0.93)
- SfM-Free 3D Gaussian Splatting via Hierarchical Training (CVPR 2025 · similarity 0.93)
- Fast Feedforward 3D Gaussian Splatting Compression (ICLR 2025 · similarity 0.93)
- FlashGS: Efficient 3D Gaussian Splatting for Large-scale and High-resolution Rendering (CVPR 2025 · similarity 0.93)
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.
