Research map / Video / Video generation
Let Your Image Move with Your Motion! -- Implicit Multi-Object Multi-Motion Transfer
Yuze Li, Dong Gong, Xiao Cao, Junchao Yuan, Dongsheng Li, Lei Zhou, Yun Sing Koh, Cheng Yan, Xinyu Zhang
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 Decoupled Mask Attention M2X masks (ablation) | FlexiMMT · Multi-motion transfer | TF: 0.381 -> 0.577 | Improved |
| Add Text-to-[X] masks (ablation) | FlexiMMT · Multi-motion transfer | TF: 0.461 -> 0.577 | Improved |
| Add training-stage mask extraction (ablation) | FlexiMMT · Multi-motion transfer | TF: 0.440 -> 0.577 | Improved |
| Add inference-stage mask extraction (ablation) | FlexiMMT · Multi-motion transfer | TF: 0.373 -> 0.577 | Improved |
| Add RMPM mask propagation at inference (ablation) | FlexiMMT · Multi-motion transfer | FF: 0.607 -> 0.723 | Improved |
| Use dynamic RMPM instead of static RMPM (ablation) | FlexiMMT · Inference time | time: 977s -> 564s | Improved |
| Add masks for appearance metrics (AC) vs removing masks (ablation) | FlexiMMT · Multi-motion transfer | AC: 0.926 -> 0.904 (vs w/o T2X) | Hurt |
| FlexiMMT vs CogVideoX-5B-I2V baseline | Image-to-video motion transfer · Multi-motion transfer | TF: 0.292 -> 0.577 | Improved |
| FlexiMMT vs best baseline in human evaluation | Image-to-video motion transfer · Human Motion Fidelity | % chosen: 6.500 -> 89.475 | Improved |
Closest papers by meaning
- Through-The-Mask: Mask-based Motion Trajectories for Image-to-Video Generation (CVPR 2025 · similarity 0.89)
- Follow-Your-Click: Open-domain Regional Image Animation via Motion Prompts (AAAI 2025 · similarity 0.87)
- TrajLoc: Trajectory-Attention Localization for Multi-Object Motion Control (NeurIPS 2026 · similarity 0.86)
- ConMo: Controllable Motion Disentanglement and Recomposition for Zero-Shot Motion Transfer (CVPR 2025 · similarity 0.86)
- DisMo: Disentangled Motion Representations for Open-World Motion Transfer (NeurIPS 2025 · similarity 0.86)
- MotionPro: A Precise Motion Controller for Image-to-Video Generation (CVPR 2025 · similarity 0.86)
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.
