Research map / Video / Video generation
MotionFlow: Attention-Driven Motion Transfer in Video Diffusion Models
Tuna Han Salih Meral, Hidir Yesiltepe, Connor Dunlop, Pinar Yanardag
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 attention-guided latent updates on top of DDIM-inverted latent (ablation) | MotionFlow · DAVIS prompts (qualitative) | motion and subject quality | Improved |
| optimize latents toward source attention masks instead of fine-tuning (vs MotionDirector) | MotionFlow · DAVIS | Text Similarity: 0.307 -> 0.322 | Improved |
| same (vs DMT) | MotionFlow · DAVIS | Motion Fidelity: 0.960 -> 0.940 | Hurt |
| same (vs Motion Inversion) | MotionFlow · DAVIS | Temporal Consistency: 0.941 vs 0.941 | No clear effect |
| same (vs DMT) | MotionFlow · DAVIS | FID: 218.99 -> 230.89 | Hurt |
| same (vs best baseline) | MotionFlow · User study | preference fraction (text alignment): 0.20 -> 0.42 | Improved |
| same (vs DMT) | MotionFlow · Processing time | initial setup time: 258 -> 49 | Improved |
| same (vs Motion Inversion) | MotionFlow · Processing time | total time: 225 -> 425 | Hurt |
Closest papers by meaning
- Video Motion Transfer with Diffusion Transformers (CVPR 2025 · similarity 0.90)
- FlowMo: Variance-Based Flow Guidance for Coherent Motion in Video Generation (NeurIPS 2025 · similarity 0.89)
- Space-Time Diffusion Features for Zero-Shot Text-Driven Motion Transfer (CVPR 2024 · similarity 0.89)
- VMC: Video Motion Customization using Temporal Attention Adaption for Text-to-Video Diffusion Models (CVPR 2024 · similarity 0.89)
- Video Diffusion Models are Training-free Motion Interpreter and Controller (NeurIPS 2024 · similarity 0.89)
- Optical-Flow Guided Prompt Optimization for Coherent Video Generation (CVPR 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.
