Research map / 3D vision / Gaussian splatting
On the Design of Mixture-of-Experts for Dynamic Gaussian Splatting
In-Hwan Jin, Hyeongju Mun, Joonsoo Kim, Kugjin Yun, Kyeongbo Kong
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 E-D3DGS as second deformation expert (MoDE) | 4DGaussians · N3V | PSNR: 30.61 -> 30.97 | Improved |
| add E-D3DGS as second deformation expert (MoDE) | Grid4D · N3V | PSNR: 31.22 -> 31.63 | Improved |
| add Grid4D as second deformation expert (MoDE) | E-D3DGS · N3V | PSNR: 31.55 -> 31.25 | Hurt |
| add a third expert E-D3DGS (N=2 to N=3) | MoE-GS · N3V | PSNR: 32.82 -> 33.23 | Improved |
| add a fourth expert 4DGaussians (N=3 to N=4) | MoE-GS · N3V | PSNR: 33.23 -> 33.27 | No clear effect |
| use volume-aware pixel router instead of pixel router (ablation) | MoE-GS · N3V | PSNR: 31.12 -> 33.23 | Improved |
| use volume-aware pixel router instead of volume router (ablation) | MoE-GS · N3V | PSNR: 32.05 -> 33.23 | Improved |
| add gate-aware Gaussian pruning (ablation) | MoE-GS (N=3) · N3V | PSNR and FPS: 32.54 -> 33.23 PSNR; 60 -> 68 FPS | Improved |
| add single-pass multi-expert rendering (ablation) | MoE-GS (N=3) · N3V | FPS: 40 -> 68 | Improved |
| prune 55% of Gaussians | MoE-GS (N=2) · N3V | PSNR and FPS: 32.82 -> 32.80 PSNR; 44 -> 83 FPS | No clear effect |
| distill MoE-GS into the single E-D3DGS expert | E-D3DGS · Technicolor | PSNR: 32.88 -> 33.67 | Improved |
| weight distillation with MoE routing weights (ablation) | E-D3DGS distillation · Technicolor | PSNR: 33.26 -> 33.67 | 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.
- Grid4D — Grid4D: 4D Decomposed Hash Encoding for High-Fidelity Dynamic Gaussian Splatting (NeurIPS 2024) · 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)
- MoDec-GS: Global-to-Local Motion Decomposition and Temporal Interval Adjustment for Compact Dynamic 3D Gaussian Splatting (CVPR 2025 · similarity 0.89)
- SC-GS: Sparse-Controlled Gaussian Splatting for Editable Dynamic Scenes (CVPR 2024 · similarity 0.89)
- Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction (CVPR 2024 · similarity 0.89)
- MotionGS: Exploring Explicit Motion Guidance for Deformable 3D Gaussian Splatting (NeurIPS 2024 · similarity 0.89)
- HAIF-GS: Hierarchical and Induced Flow-Guided Gaussian Splatting for Dynamic Scene (NeurIPS 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.
