Research map / 3D vision / Gaussian splatting
Dynamic Gaussian Scene Reconstruction from Unsynchronized Videos
Zhixin Xu, Hengyu Zhou, Yuan Liu, Wenhan Xue, Hao Pan, Wenping Wang, Bin Wang
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 coarse temporal alignment (ablation) | 4DGaussians · DyNeRF unsynchronized | PSNR: 29.56 -> 30.92 | Improved |
| add fine temporal refinement (ablation) | 4DGaussians · DyNeRF unsynchronized | PSNR: 29.56 -> 30.87 | Improved |
| add fine refinement on top of coarse alignment (ablation) | 4DGaussians · DyNeRF unsynchronized | PSNR: 30.92 -> 31.16 | Improved |
| add coarse-to-fine alignment (ablation) | 4DGaussians · DyNeRF unsynchronized | LPIPS: 0.099 -> 0.091 | Improved |
| add alignment module at max offset 3 (ablation) | 4DGaussians · DyNeRF random offset 3 | PSNR: 30.69 -> 31.25 | Improved |
| add alignment module at max offset 5 (ablation) | 4DGaussians · DyNeRF random offset 5 | PSNR: 30.31 -> 31.29 | Improved |
| add alignment module at max offset 10 (ablation) | 4DGaussians · DyNeRF random offset 10 | PSNR: 29.60 -> 31.16 | Improved |
| increase max time offset from 3 to 10 frames (ablation) | 4DGaussians baseline · DyNeRF | PSNR: 30.69 -> 29.60 | Hurt |
| increase max time offset from 3 to 10 frames (ablation) | 4DGaussians+Ours · DyNeRF | PSNR: 31.25 -> 31.16 | No clear effect |
| add alignment module | SC-GS · DyNeRF Coffee Martini | PSNR: 26.81 -> 25.68 | Hurt |
| add alignment module | RT4DGS · DyNeRF Cook Spinach | PSNR: 31.15 -> 33.15 | Improved |
| add alignment module | 4DGaussians · DyNeRF Sear Steak | PSNR: 29.67 -> 32.51 | 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.
- SC-GS — SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes (CVPR 2024) · better on 6 of 6 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
- 4D Gaussian Splatting in the Wild with Uncertainty-Aware Regularization (NeurIPS 2024 · similarity 0.90)
- Instant4D: 4D Gaussian Splatting in Minutes (NeurIPS 2025 · similarity 0.90)
- DGS-LRM: Real-Time Deformable 3D Gaussian Reconstruction From Monocular Videos (NeurIPS 2025 · similarity 0.90)
- Motion Matters: Compact Gaussian Streaming for Free-Viewpoint Video Reconstruction (NeurIPS 2025 · similarity 0.90)
- Orientation-anchored Hyper-Gaussian for 4D Reconstruction from Casual Videos (NeurIPS 2025 · similarity 0.90)
- HiCoM: Hierarchical Coherent Motion for Dynamic Streamable Scenes with 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.
