Research map / Video / Video generation
Reangle-A-Video: 4D Video Generation as Video-to-Video Translation
Hyeonho Jeong, Suhyeon Lee, Jong Chul Ye
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 stochastic control guidance (MEt3R reward) to multi-view image inpainting (ablation) | Reangle-A-Video · Multi-view warped image inpainting | MEt3R: 0.1431 -> 0.1184 | Improved |
| add stochastic control guidance (ablation) | Reangle-A-Video · Multi-view warped image inpainting | SED: 1.1966 -> 1.1844 | Improved |
| add stochastic control guidance (ablation) | Reangle-A-Video · Multi-view warped image inpainting | TSED: 0.5241 -> 0.5588 | Improved |
| add warped videos to fine-tuning data (ablation) | Reangle-A-Video · Multi-view motion fidelity (user study) | preference rate: 19.56% -> 80.44% | Improved |
| add masked diffusion loss during fine-tuning (ablation) | Reangle-A-Video · qualitative | artifacts | Improved |
| compare against vanilla CogVideoX I2V in static view transport | Reangle-A-Video · Static view transport | MEt3R: 0.0539 -> 0.0412 | Improved |
| compare against vanilla CogVideoX I2V in static view transport | Reangle-A-Video · Static view transport | FID: 79.621 -> 53.448 | Improved |
| compare against Generative Camera Dolly | Reangle-A-Video · Static view transport | FVD: 5264.7 -> 2690.9 | Improved |
| compare against NVS-Solver | Reangle-A-Video · Dynamic camera control | MEt3R: 0.1090 -> 0.0648 | Improved |
| compare against Trajectory Attention | Reangle-A-Video · Dynamic camera control | FID: 109.20 -> 74.194 | Improved |
| compare against vanilla CogVideoX on imaging quality and background consistency | Reangle-A-Video · Static view transport | VBench imaging quality: 0.6475 -> 0.6414 | No clear effect |
| compare against Trajectory Attention on dynamic degree | Reangle-A-Video · Dynamic camera control | VBench dynamic degree: 0.8889 -> 0.8884 | No clear effect |
Closest papers by meaning
- Vivid-ZOO: Multi-View Video Generation with Diffusion Model (NeurIPS 2024 · similarity 0.88)
- SV4D: Dynamic 3D Content Generation with Multi-Frame and Multi-View Consistency (ICLR 2025 · similarity 0.88)
- Reconstruct, Inpaint, Test-Time Finetune: Dynamic Novel-view Synthesis from Monocular Videos (NeurIPS 2025 · similarity 0.88)
- 4Diffusion: Multi-view Video Diffusion Model for 4D Generation (NeurIPS 2024 · similarity 0.88)
- ViVid-1-to-3: Novel View Synthesis with Video Diffusion Models (CVPR 2024 · similarity 0.88)
- SynCamMaster: Synchronizing Multi-Camera Video Generation from Diverse Viewpoints (ICLR 2025 · similarity 0.88)
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.
