AI Paper Atlas

Research map / Video / Video generation

LiveGesture: Streamable Co-Speech Gesture Generation Model

Muhammad Usama Saleem, Mayur Jagdishbhai Patel, Ekkasit Pinyoanuntapong, Zhongxing Qin, Li Yang, Hongfei Xue, Ahmed Helmy, Chen Chen, Pu Wang

CVPR 2026 · Video › Video generation · approach: Talking face generation · 2 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 temporal attention in fusion transformer (ablation)LiveGesture · BEAT2FGD: 15.52 -> 4.57Improved
add spatial attention in fusion transformer (ablation)LiveGesture · BEAT2FGD: 6.64 -> 4.57Improved
add PILOR low-rank pre-infusion alignment (ablation)LiveGesture · BEAT2FGD: 4.89 -> 4.57Improved
add uncertainty-guided refinement (UGR) (ablation)LiveGesture · BEAT2BC: 0.723 -> 0.794Improved
add text cues (ablation)LiveGesture · BEAT2FGD: 4.60 -> 4.57No clear effect
add xAR-Fuse fusion on top of region eXperts (ablation)Local region eXperts · BEAT2FGD: 6.458 -> 4.57Improved
two-stage SVQ with only Quant+MLP trainable instead of 1-stage joint training (ablation)SVQ tokenizer · BEAT2FGD: 4.691 -> 4.557Improved
increase codebook size from 512 to 2048 (ablation)SVQ tokenizer · BEAT2FGD: 6.63 -> 4.51Improved
use causal cross-attention instead of additive causal self-attention (ablation)HAR · BEAT2FGD: 4.63 -> 4.57Improved
use lightweight causal audio encoder instead of Magic-Codec (ablation)audio encoder · BEAT2FGD: 4.51 -> 4.57No clear effect
set local expert loss weight 0.3 instead of 0.0 (ablation)LiveGesture · BEAT2FGD: 4.89 -> 4.57Improved
set local expert loss weight 1.0 instead of 0.3 (ablation)LiveGesture · BEAT2FGD: 4.57 -> 4.85Hurt

Earlier methods this paper compares with

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