Research map / 3D vision / Gaussian splatting
SaLon3R: Structure-Aware Long-Term Feedforward 3D Reconstruction from Unposed Images
Jiaxin Guo, Tongfan Guan, Wenzhen Dong, Wenzhao Zheng, Wenting Wang, Yue Wang, Yeung Yam, Yun‐Hui Liu
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 plain quantization (ablation) | baseline Gaussian head · ScanNet component ablation | PSNR: 20.78 -> 16.95 | Hurt |
| add plain quantization (ablation) | baseline Gaussian head · ScanNet component ablation | Gaussian count (K): 1966 -> 616 | Improved |
| make quantization saliency-guided (ablation) | + Quant. · ScanNet component ablation | PSNR: 16.95 -> 18.06 | Improved |
| add GS refiner (point transformer) (ablation) | + Sal. Quant. · ScanNet component ablation | PSNR: 18.06 -> 21.65 | Improved |
| add adaptive Gaussian growing (ablation) | + GS Refiner · ScanNet component ablation | PSNR: 21.65 -> 22.45 | Improved |
| add edge-aware smoothness loss (ablation) | + Adapt. Grow · ScanNet component ablation | Abs Rel: 0.078 -> 0.030 | Improved |
| use VGGT backbone instead of CUT3R (ablation) | SaLon3R · ScanNet novel view synthesis (10 views) | PSNR: 22.54 -> 23.26 | Improved |
| increase context views from 10 to 30 (ablation) | SaLon3R · Replica office02 context views | LPIPS: 0.429 -> 0.317 | Improved |
| increase context views from 10 to 100 (ablation) | SaLon3R · Replica office02 context views | PSNR: 25.13 -> 23.05 | Hurt |
| increase context views from 10 to 200 (ablation) | SaLon3R · Replica office02 context views | FPS: 10.23 -> 9.53 | No clear effect |
| increase quantization resolution from 0.01 to 0.02 (ablation) | SaLon3R · Replica office02 quantization resolution | Abs Rel: 0.050 -> 0.092 | Hurt |
| apply quantization at resolution 0.01 instead of none (ablation) | SaLon3R · Replica office02 quantization resolution | Gaussian count (K): 5629 -> 2005 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- FreeSplat — FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024) · better on 9 of 11 shared results
- FLARE — FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Views (CVPR 2025) · better on 1 of 1 shared result
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
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.91)
- No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images (ICLR 2025 · similarity 0.89)
- FreeSplat: Generalizable 3D Gaussian Splatting Towards Free View Synthesis of Indoor Scenes (NeurIPS 2024 · similarity 0.89)
- Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos (NeurIPS 2025 · similarity 0.89)
- 4D3R: Motion-Aware Neural Reconstruction and Rendering of Dynamic Scenes from Monocular Videos (NeurIPS 2025 · similarity 0.89)
- MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 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.
