Research map / 3D vision / Gaussian splatting
4DSurf: High-Fidelity Dynamic Scene Surface Reconstruction
Renjie Wu, Hongdong Li, Jose M. Alvarez, Miaomiao 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 SDF Flow Regularization (ablation) | Gaussian Velocity Field model · Hi4D (avg of 6 scenes) | Overall Chamfer Distance: 1.49 -> 1.02 | Improved |
| add independent segment partitioning (ablation) | GVF + SF-Reg · Hi4D (avg of 6 scenes) | Overall Chamfer Distance: 1.02 -> 0.77 | Improved |
| use overlapping instead of independent segments (ablation) | GVF + SF-Reg · Hi4D (avg of 6 scenes) | Overall Chamfer Distance: 0.77 -> 0.67 | Improved |
| add Incremental Motion Tuning (LoRA rank 64) (ablation) | 4DSurf · Hi4D (avg of 6 scenes) | Overall Chamfer Distance: 0.67 -> 0.70 | No clear effect |
| lower LoRA rank 64 to 16 (ablation) | IMT · Hi4D (avg of 6 scenes) | Overall Chamfer Distance: 0.70 -> 0.89 | Hurt |
| lower LoRA rank 64 to 32 (ablation) | IMT · Hi4D (avg of 6 scenes) | Overall Chamfer Distance: 0.70 -> 0.82 | Hurt |
| use IMT rank 16 instead of full velocity fields (ablation) | 4DSurf · Hi4D Backhug02 | velocity field storage: 57.7 MB -> 9.4 MB | Improved |
| use 4DSurf instead of Dynamic-2DGS | dynamic surface reconstruction · Hi4D temporal stability | Overall STD: 1.19 -> 0.18 | Improved |
| use 4DSurf instead of Dynamic-2DGS | dynamic surface reconstruction · CMU Panoptic Band1 | Overall Chamfer Distance: 16.0 -> 12.7 | Improved |
| use 4DSurf instead of Dynamic-2DGS | dynamic surface reconstruction · Hi4D Cheers37 | Overall Chamfer Distance: 2.29 -> 0.47 | Improved |
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.
- Sparse2DGS — Sparse2DGS: Geometry-Prioritized Gaussian Splatting for Surface Reconstruction from 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
- Motion Decoupled 3D Gaussian Splatting for Dynamic Object Representation (AAAI 2025 · similarity 0.91)
- DySurface: Consistent 4D Surface Reconstruction via Bridging Explicit Gaussians and Implicit Functions (NeurIPS 2026 · similarity 0.91)
- 4D Gaussian Splatting in the Wild with Uncertainty-Aware Regularization (NeurIPS 2024 · similarity 0.91)
- HAIF-GS: Hierarchical and Induced Flow-Guided Gaussian Splatting for Dynamic Scene (NeurIPS 2025 · similarity 0.91)
- Swift4D: Adaptive divide-and-conquer Gaussian Splatting for compact and efficient reconstruction of dynamic scene (ICLR 2025 · similarity 0.91)
- Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction (CVPR 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.
