AI Paper Atlas

Research map / Video / Video generation

ExpertEdit: Learning Skill-Aware Motion Editing from Expert Videos

Arjun Somayazulu, Kristen Grauman

ECCV 2026 · Video › Video generation · approach: Trajectory-guided motion control · 1 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
use expert-manifold kinematic-masked infilling instead of text-driven supervised editorSimMotionEdit · Ego-Exo4D Rev. LayupFID improvement F (%): 4.88 -> 12.08Improved
use ExpertEdit instead of SimMotionEdit on penalty kickSimMotionEdit · Ego-Exo4D Pen. KickPose improvement P (%): 3.20 -> 6.03Improved
use ExpertEdit instead of SimMotionEdit on Rev. PunchSimMotionEdit · Kyokushin Karate Rev. PunchPose improvement P (%): 2.20 -> 2.07Hurt
use ExpertEdit instead of SimMotionEdit on High RHSimMotionEdit · Kyokushin Karate High RHPose improvement P (%): 3.30 -> 2.96Hurt
use ExpertEdit instead of SimMotionEdit on karate Front kickSimMotionEdit · Kyokushin Karate FrontFID improvement F (%): 5.05 -> 9.73Improved
prompt asking for smoother, controlled motion instead of expert-like (ablation)SimMotionEdit · Ego-Exo4D Rev. LayupFID improvement F (%): 0.81 -> 2.90Improved
add technique name to smooth prompt (ablation)SimMotionEdit · Ego-Exo4D Pen. KickPose improvement P (%): 1.74 -> 3.20Improved
increase training expert data from 30% to 100% (ablation)ExpertEdit · Average over techniquesP and FImproved
halve mask span fraction alpha 0.3 -> 0.15 (ablation)ExpertEdit · Average over techniquesPose improvement P (%): 4.02 -> 4.09; F 7.01 -> 6.10Mixed
increase mask span fraction alpha 0.3 -> 0.45 (ablation)ExpertEdit · Average over techniquesFID improvement F (%): 7.01 -> 6.41Hurt
ExpertEdit vs SimMotionEdit in human preferenceExpertEdit · Human preference (basketball+soccer)preferred %: 43.8% vs 25.9%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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