AI Paper Atlas

Research map / Video / Video generation

Time-to-Move: Training-Free Motion-Controlled Video Generation via Dual-Clock Denoising

Assaf Singer, Noam Rotstein, Amir Mann, Ron Kimmel, Or Litany

ICLR 2026 · Video › Video generation · approach: Trajectory-guided motion control · 10 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 dual clock (t_weak, t_strong) instead of single clock at t_weak (ablation)TTM · MC-Bench (SVD)CoTracker distance: 27.316 -> 7.967Improved
use dual clock instead of single clock at t_strong (ablation)TTM · MC-Bench (SVD)dynamic degree: 0.353 -> 0.427Improved
use dual clock instead of single clock at t_strong (ablation)TTM · MC-Bench (SVD)CoTracker distance: 5.528 -> 7.967Hurt
use dual clock instead of RePaint-style second tick 0 (ablation)TTM · MC-Bench (SVD)imaging quality: 0.576 -> 0.617Improved
use dual clock instead of RePaint-style second tick 0 (ablation)TTM · MC-Bench (SVD)CoTracker distance: 2.923 -> 7.967Hurt
start first tick at t_weak instead of pure noise T (ablation)TTM · MC-Bench (SVD)CoTracker distance: 9.228 -> 7.967Improved
start first tick at t_weak instead of pure noise T (ablation)TTM · MC-Bench (SVD)dynamic degree: 0.430 -> 0.427No clear effect
start from pure noise T with second tick t_weak (ablation)TTM · MC-Bench (SVD)CoTracker distance: 29.399Hurt
use TTM instead of GWTFGWTF · MC-Bench (CogVideoX)CoTracker distance: 27.844 -> 13.665Improved
use TTM instead of GWTFGWTF · DL3DVMSE: 0.033 -> 0.022Improved
use TTM instead of GWTFGWTF · MC-Bench (CogVideoX)dynamic degree: 0.764 -> 0.357Hurt
use TTM instead of depth-warped reference onlyWarped · DL3DVFID: 33.443 -> 21.966Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 3 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