Research map / Video / Video generation
UniScene-MoTion: Unified Scene & Motion-aware Diffusion Transition Framework
Rui Jiang, Chongmian Wang, Xinghe Fu, Yehao Lu, Teng Li, Xi Li
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 progressive training with disparity supervision (ablation) | UniScene-MoTion · VBench RealEstate10K (81 frames) | Subject Consistency / Image Quality: SC 0.926 -> 0.939; IQ 0.72 -> 0.729 | Improved |
| add bidirectional interactive feature injection (ablation) | UniScene-MoTion · VBench RealEstate10K (81 frames) | Subject Consistency / Image Quality: SC 0.929 -> 0.939; IQ 0.708 -> 0.729 | Improved |
| add bidirectional injection and progressive training (ablation) | UniScene-MoTion · VBench motion smoothness / dynamic degree | Motion Smoothness, Dynamic Degree: MS 0.995/0.994 -> 0.994; DD 0.698/0.692 -> 0.690 | No clear effect |
| UniScene-MoTion vs FCVG on 25-frame transitions | video transition · RealEstate10K (25 frames) | FID: 3.254 -> 2.415 | Improved |
| UniScene-MoTion vs FCVG on 25-frame transitions | video transition · RealEstate10K (25 frames) | FVD: 125.928 -> 107.133 (SEINE 106.962 is best) | Hurt |
| UniScene-MoTion (81 frames) vs VideoX-Fun 14B | video transition · RealEstate10K (81 frames) | FVMD / FVD: FVMD 2135.053 -> 227.617; FVD 218.279 -> 133.213 | 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.
- SEINE — SEINE: Short-to-Long Video Diffusion Model for Generative Transition and Prediction (ICLR 2024) · better on 3 of 4 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
- Seurat: From Moving Points to Depth (CVPR 2025 · similarity 0.88)
- AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers (CVPR 2025 · similarity 0.87)
- Frame In-N-Out: Unbounded Controllable Image-to-Video Generation (NeurIPS 2025 · similarity 0.87)
- VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control (ICLR 2025 · similarity 0.87)
- Emergent Temporal Correspondences from Video Diffusion Transformers (NeurIPS 2025 · similarity 0.87)
- Diffusion4D: Fast Spatial-temporal Consistent 4D generation via Video Diffusion Models (NeurIPS 2024 · similarity 0.87)
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.
