Research map / 3D vision / Gaussian splatting
Uncertainty Matters in Dynamic Gaussian Splatting for Monocular 4D Reconstruction
Fengzhi Guo, Chih-Chuan Hsu, Sihao Ding, Cheng Zhang
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 uncertainty-based key node selection (vs uniform 2D sampling) (ablation) | USplat4D on MoSca · DyCheck (7 scenes, 2x) | mPSNR: 18.86 -> 19.63 | Improved |
| add uncertainty-aware kNN edges (vs distance-only kNN) (ablation) | USplat4D on MoSca · DyCheck (7 scenes, 2x) | mPSNR: 19.50 -> 19.63 | Improved |
| add uncertainty weighting in the loss (ablation) | USplat4D on MoSca · DyCheck (7 scenes, 2x) | mPSNR: 19.08 -> 19.63 | Improved |
| add 3D gridization for key node candidates (vs random sampling) (ablation) | USplat4D on MoSca · DyCheck (7 scenes, 2x) | mPSNR: 19.50 -> 19.63 | Improved |
| add uncertainty-aware graph optimization to SoM | SoM · DyCheck (5 scenes, 1x) | mPSNR: 16.72 -> 16.85 | Improved |
| add uncertainty-aware graph optimization to MoSca | MoSca · DyCheck (7 scenes, 2x) | mPSNR: 19.32 -> 19.63 | Improved |
| add uncertainty-aware graph optimization to SoM | SoM · Objaverse (views 120-180 deg) | PSNR: 16.45 -> 17.03 | Improved |
| add uncertainty-aware graph optimization to MoSca | MoSca · Objaverse (views 120-180 deg) | PSNR: 15.89 -> 16.31 | Improved |
| add uncertainty-aware graph optimization to SoM | SoM · Objaverse (views 120-180 deg) | LPIPS: 0.31 -> 0.26 | Improved |
| add uncertainty-aware graph optimization to SoM | SoM · DyCheck tracking (SoM protocol) | PCK 5cm: 43.0 -> 54.4 | Improved |
| add uncertainty-aware graph optimization to MoSca | MoSca · DyCheck tracking (MoSca protocol) | PCK 5%: 82.4 -> 84.5 | Improved |
| add uncertainty-aware graph optimization to MoSca | MoSca · Objaverse Panda scene (views 0-60 deg) | PSNR: 15.37 -> 15.08 | Hurt |
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.
- MoSca — MoSca: Dynamic Gaussian Fusion from Casual Videos via 4D Motion Scaffolds (CVPR 2025) · better on 5 of 5 shared results
- MoDec-GS — MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 3D Gaussian Splatting (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
- 4D Gaussian Splatting in the Wild with Uncertainty-Aware Regularization (NeurIPS 2024 · similarity 0.94)
- Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos (NeurIPS 2025 · similarity 0.91)
- Fully Explicit Dynamic Gaussian Splatting (NeurIPS 2024 · similarity 0.91)
- Motion Decoupled 3D Gaussian Splatting for Dynamic Object Representation (AAAI 2025 · similarity 0.91)
- Gaussian Splatting Lucas-Kanade (ICLR 2025 · similarity 0.91)
- Variational Multi-scale Representation for Estimating Uncertainty in 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.90)
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.
