AI Paper Atlas

Research map / Video / Video generation

MotionFlow: Attention-Driven Motion Transfer in Video Diffusion Models

Tuna Han Salih Meral, Hidir Yesiltepe, Connor Dunlop, Pinar Yanardag

AAAI 2026 · Video › Video generation · approach: Efficient video diffusion models · 17 citations (OpenAlex)

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 changedWhereMeasuredEffect
add attention-guided latent updates on top of DDIM-inverted latent (ablation)MotionFlow · DAVIS prompts (qualitative)motion and subject qualityImproved
optimize latents toward source attention masks instead of fine-tuning (vs MotionDirector)MotionFlow · DAVISText Similarity: 0.307 -> 0.322Improved
same (vs DMT)MotionFlow · DAVISMotion Fidelity: 0.960 -> 0.940Hurt
same (vs Motion Inversion)MotionFlow · DAVISTemporal Consistency: 0.941 vs 0.941No clear effect
same (vs DMT)MotionFlow · DAVISFID: 218.99 -> 230.89Hurt
same (vs best baseline)MotionFlow · User studypreference fraction (text alignment): 0.20 -> 0.42Improved
same (vs DMT)MotionFlow · Processing timeinitial setup time: 258 -> 49Improved
same (vs Motion Inversion)MotionFlow · Processing timetotal time: 225 -> 425Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy