AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

Expressive Talking Head Generation With Granular Audio-Visual Control

Borong Liang, Yan Pan, Zhizhi Guo, Hang Zhou, Zhibin Hong, Xiaoguang Han, Junyu Han, Jingtuo Liu, Errui Ding, Jingdong Wang

CVPR 2022 · Vision-language and multimodal › Cross-modal learning · approach: Audio-visual speech separation · 119 citations (OpenAlex)

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 the granular audio-driven pose, expression, and speech-content controlsGC-AVT · MEADLMDm: 4.014 -> 3.252Improved
Add the granular audio-driven pose, expression, and speech-content controlsGC-AVT · VoxCeleb2LMDm: 3.662 -> 3.356Improved
Add independent pose, expression, and visual speech-content controlGC-AVT · MEADLMDm: 2.966 -> 2.349Improved
Add independent pose, expression, and visual speech-content controlGC-AVT · VoxCeleb2LMDm: 4.035 -> 2.811Improved
Add VGG and VGGFace perceptual losses (ablation)GC-AVT · VoxCeleb2LMD: 4.753 -> 3.025Improved
Add VGG and VGGFace perceptual losses (ablation)GC-AVT · VoxCeleb2SSIM: 0.662 -> 0.710Improved
Add contrastive audio-visual synchronization loss (ablation)GC-AVT · VoxCeleb2Syncconf: 4.066 -> 5.250Improved
Add contrastive audio-visual synchronization loss (ablation)GC-AVT · VoxCeleb2LMDm: 4.311 -> 3.356Improved
Add the time-shifted mouth input to expression processing (ablation)GC-AVT · VoxCeleb2LMDm: 3.704 -> 3.356Improved
Add the time-shifted mouth input to expression processing (ablation)GC-AVT · VoxCeleb2Syncconf: 4.760 -> 5.250Improved
Remove time-shifted mouth input from expression processing (ablation)GC-AVT · VoxCeleb2SSIM: 0.684 -> 0.710No clear effect
Use the proposed masking design rather than the tested alternative mask layouts (ablation)GC-AVT · VoxCeleb2speech-content controlImproved

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