AI Paper Atlas

Research map / Video / Video generation

Learning to Generate Rigid Body Interactions with Video Diffusion Models

David Romero, Ariana Bermudez, Viacheslav Iablochnikov, Hao Li, Fabio Pizzati, Ivan Laptev

ECCV 2026 · Video › Video generation · approach: Trajectory-guided motion control · 7 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 first-stage training with full masks before second-stage dropout (ablation)KineMask · Simple MotionFVD: 288.8 -> 160.3Improved
add first-stage training with full masks (ablation)KineMask · Simple MotionIoU: 0.237 -> 0.367Improved
train KineMask vs zero-shot CogVideoXCogVideoX · InteractionsFVD: 807.3 -> 250.7Improved
train on Interactions instead of Simple Motion (ablation)KineMask · InteractionsFVD: 301.1 -> 250.7Improved
KineMask on Simple Motion vs plain CogVideoX finetuned on InteractionsCogVideoX · InteractionsFVD: 368.2 -> 301.1Improved
train with rich captions c instead of generic prompt (ablation)KineMask · InteractionsIoU / FVD / FMVD: IoU 0.356 -> 0.376; FVD 238.8 -> 231.3; FMVD 161.3 -> 174.4Mixed
apply KineMask to Wan2.2-5BWan2.2-5B · InteractionsFVD: 977.7 -> 475.6Improved
apply KineMask to Cosmos2.5-2BCosmos2.5-2B · InteractionsFVD: 1287.3 -> 433.6Improved
encode per-frame instantaneous velocity v_t instead of v0 (ablation)KineMask · InteractionsFVD: 277.3 -> 231.3Improved
use non-uniform dropout sampling biased to interaction frames (ablation)KineMask · InteractionsFVD: 303.1 -> 231.3Improved
KineMask vs finetuned CogVideoX on real Physics-IQCogVideoX · Physics-IQFVD: 357.1 -> 212.1Improved
KineMask vs Force-Prompting on Physics-IQForce-Prompting · Physics-IQMSE: 27.76 -> 39.67Hurt

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