AI Paper Atlas

Research map / Video / Video generation

A Plug-and-Play Physical Motion Restoration Approach for In-the-Wild High-Difficulty Motions

Youliang Zhang, Ronghui Li, Yachao Zhang, Liang Pan, Jingbo Wang, Yebin Liu, Xiu Li

ICCV 2025 · Video › Video generation · approach: Trajectory-guided motion control · 4 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 MCM and physics fine-tuning with mask guidance (ablation)PTM pipeline · High-difficulty in-the-wild datasetsuccess rate: 65% -> 87%Improved
use human masks instead of 2D keypoints as guidance (ablation)MCM and PTM · High-difficulty in-the-wild datasetOKS, MPS, SRImproved
add motion prior initialization (ablation)PTM adaptation · High-difficulty in-the-wild datasetsuccess rate: 77% -> 87%Improved
add motion prior and residual action (ablation)PTM adaptation · High-difficulty in-the-wild datasetaverage adaptation steps: 4.5k -> 2.5kImproved
replace PHC+ with pretrained PTM before adaptation (ablation)PHC+ · High-difficulty in-the-wild datasetsuccess rate: 21% -> 37%Improved
add the plug-and-play restoration to TRAMTRAM · High-difficulty in-the-wild datasetground penetration: 19.988 -> 0.595Improved
add the plug-and-play restoration to GVHMRGVHMR · High-difficulty in-the-wild datasetOKS: 0.837 -> 0.865Improved
add PhysPT post-processing to GVHMRGVHMR · High-difficulty in-the-wild datasetOKS: 0.837 -> 0.806Hurt
add restoration to T2M-GPTT2M-GPT · HumanML3DFID: 0.116 -> 0.119No clear effect
add restoration to MomaskMomask · HumanML3DFID: 0.045 -> 0.062Hurt
add restoration to MomaskMomask · HumanML3Dground penetration: 20.672 -> 0.233Improved
use PTM instead of PHC+ for imitationPHC+ · AIST++ and Kungfusuccess rate: 72.74% -> 96.87%Improved

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