AI Paper Atlas

Research map / Video / Video generation

FastVMT: Eliminating Redundancy in Video Motion Transfer

Yue Ma, Zhikai Wang, Tianhao Ren, Mingzhe Zheng, Hongyu Liu, Jiayi Guo, Kunyu Feng, Yuxuan Xue, Zixiang Zhao, Konrad Schindler, Qifeng Chen, Linfeng Zhang

ICLR 2026 · Video › Video generation · approach: Efficient video diffusion models · 29 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 sliding-window motion extraction (ablation)FastVMT · VBench ablationbackground consistency: 0.9437 -> 0.9684Improved
add corresponding-window loss (ablation)FastVMT · VBench ablationmotion smoothness: 0.9705 -> 0.9891Improved
add step-skipping gradient reuse (ablation)FastVMT · VBench ablationaesthetic quality: 0.5623 -> 0.5778Improved
add step-skipping gradient reuse (ablation)FastVMT · VBench ablationsubject consistency: 0.9852 -> 0.9809No clear effect
extract motion from layer 15 rather than 10 or 20 (ablation)FastVMT · DAVISmotion fidelity: 0.7128 / 0.7213 -> 0.7471Improved
use temporal span 5 rather than 3 or 7 (ablation)FastVMT · VBench ablationsubject consistency: 0.9592 / 0.9711 -> 0.9809Improved
use FastVMT instead of MotionClone (ablation)training-free motion transfer · DAVIStime (s): 397.05 -> 184.18Improved
use FastVMT instead of DiTFlowtraining-free motion transfer · DAVISmotion fidelity: 0.4062 -> 0.7471Improved

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.

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

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