AI Paper Atlas

Research map / Video / Video generation

DanceTogether: Generating Interactive Multi-Person Video without Identity Drifting

Junhao Chen, Mingjin Chen, Jianjin Xu, Xiang Li, Junting Dong, Mingze Sun, Puhua Jiang, Hongxiang Li, Yuhang Yang, Hao Zhao, Xiao-Xiao Long, Ruqi Huang

ICLR 2026 · Video › Video generation · approach: Diffusion-based human image animation

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 MaskPoseAdapter instead of plain PoseNet (ablation)DanceTog · DanceTogEval-100HOTA: 48.95 -> 83.94Improved
replace attention fusion with simple summation (ablation)DanceTog · DanceTogEval-100HOTA: 83.94 -> 58.23Hurt
add separate mask input (ablation)DanceTog · DanceTogEval-100IDF1: 42.49 -> 89.59Improved
add separate pose input (ablation)DanceTog · DanceTogEval-100MPJPE2D: 1292.33 -> 492.24Improved
add MultiFaceEncoder (ablation)DanceTog · DanceTogEval-100MPJPE2D: 893.32 -> 492.24Improved
add PairFS-4K training data (ablation)DanceTog · DanceTogEval-100HOTA: 81.79 -> 83.94Improved
add all multi-person datasets beyond Swing Dance (ablation)DanceTog · DanceTogEval-100HOTA: 80.26 -> 81.79Improved
use DanceTog architecture instead of StableAnimator, both on Swing DanceDanceTog · DanceTogEval-100HOTA: 71.35 -> 80.26Improved
fine-tune StableAnimator on Swing Dance dataStableAnimator · DanceTogEval-100HOTA: 67.75 -> 71.35Improved
DanceTog vs StableAnimator w. Dataswing on mask-region FVDDanceTog · DanceTogEval-100 (human masked region)FVD (masked): 29.0 -> 17.1Improved
DanceTog vs StableAnimator inference speedDanceTog · DanceTogEval-100FPS: 0.89 -> 0.88No clear effect

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