AI Paper Atlas

Research map / Video / Video generation

ActAvatar: Temporally-Aware Precise Action Control for Talking Avatars

Ziqiao Peng, Yi Chen, Yifeng Ma, Guozhen Zhang, Zhiyao Sun, Zixiang Zhou, Youliang Zhang, Zhengguang Zhou, Zhaoxin Fan, Hongyan Liu, Yuan Zhou, Qinglin Lu and 1 more

CVPR 2026 · Video › Video generation · approach: Talking face generation · 3 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 Phase-Aware Cross-Attention (PACA) (ablation)base global-prompt model · Action BenchHit@Segment: 0.725 -> 0.829Improved
add PACA (ablation)base global-prompt model · Action BenchAction Accuracy: 3.91 -> 5.78Improved
add PACA (ablation)base global-prompt model · Action BenchTemporal Correctness: 6.47 -> 7.12Improved
add PACA (ablation)base global-prompt model · Action BenchSync-C: 6.37 -> 6.39No clear effect
add progressive audio-visual alignment (ablation)PACA model · Action BenchSync-C: 6.39 -> 6.57Improved
add progressive audio-visual alignment (ablation)PACA model · Action BenchHit@Segment: 0.829 -> 0.831No clear effect
add two-stage training (ablation)PACA + progressive alignment · Action BenchSync-C: 6.57 -> 6.89Improved
add two-stage training (ablation)PACA + progressive alignment · Action BenchHit@Segment: 0.831 -> 0.854Improved
add two-stage training (ablation)PACA + progressive alignment · Action BenchAction Accuracy: 5.75 -> 5.97Improved
ActAvatar vs best baseline on lip syncActAvatar · HDTF test setSync-C: 7.663 vs 7.986 (OmniAvatar)No clear effect
ActAvatar vs Wan-S2V on visual qualityActAvatar · HDTF test setFID: 23.850 -> 23.471Improved
ActAvatar with 5B params vs 14B modelsActAvatar · HDTF test setinference time: 68 min (Wan-S2V) -> 16 minImproved

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