Research map / Video / Video generation
Diff4Splat: Repurposing Video Diffusion Models for Dynamic Scene Generation
Panwang Pan, Chenguo Lin, Chenxin Li, Jingjing Zhao, Yuchen Lin, Haopeng Li, Yunlong Lin, Kairun Wen, Yixuan Yuan, Yadong MU
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 motion loss (ablation) | Diff4Splat · single-image dynamic scene generation (160-sample benchmark) | FVD: 351.382 -> 210.153 | Improved |
| add motion loss (ablation) | Diff4Splat · single-image dynamic scene generation (160-sample benchmark) | QA-Quality: 2.145 -> 2.813 | Improved |
| add motion loss (ablation) | Diff4Splat · single-image dynamic scene generation (160-sample benchmark) | Avg. Matches: 4821.56 -> 5114.22 | Improved |
| add deformation Gaussian field module (ablation) | Diff4Splat · qualitative | visual quality | Improved |
| add static pre-training before dynamic training (progressive training) (ablation) | Diff4Splat · qualitative, 100K iterations | visual quality and training time: 21 vs 7 days for comparable baseline | Improved |
| use explicit 3D Gaussian field instead of implicit video generation | AC3D · pose fidelity | avg RPE translation: 3.001 -> 0.012 | Improved |
| use explicit 3D Gaussian field instead of implicit video generation | AC3D · pose fidelity | avg RPE rotation: 0.810 -> 0.008 | Improved |
| use feed-forward Diff4Splat instead of AC3D + Mosca | two-stage optimization pipeline · single-image dynamic scene generation (160-sample benchmark) | FVD: 235.961 -> 210.153 | Improved |
| use feed-forward Diff4Splat instead of AC3D + Mosca | two-stage optimization pipeline · single-image dynamic scene generation (160-sample benchmark) | KVD: 2.012 -> 2.316 | Hurt |
| use feed-forward Diff4Splat instead of AC3D + Mosca | two-stage optimization pipeline · single-image dynamic scene generation (160-sample benchmark) | QA-Quality: 2.842 -> 2.813 | Mixed |
Closest papers by meaning
- Diffusion4D: Fast Spatial-temporal Consistent 4D generation via Video Diffusion Models (NeurIPS 2024 · similarity 0.92)
- 4Diffusion: Multi-view Video Diffusion Model for 4D Generation (NeurIPS 2024 · similarity 0.92)
- 4Real: Towards Photorealistic 4D Scene Generation via Video Diffusion Models (NeurIPS 2024 · similarity 0.91)
- Diffusion$^2$: Dynamic 3D Content Generation via Score Composition of Video and Multi-view Diffusion Models (ICLR 2025 · similarity 0.91)
- SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency (ICLR 2025 · similarity 0.91)
- A 3D Scene is Worth 1K Tokens: 3D-Grounded Representation for Scene Generation at Scale (NeurIPS 2026 · 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.
