AI Paper Atlas

Research map / Video / Video generation

Talking Head Generation with Probabilistic Audio-to-Visual Diffusion Priors

Zhentao Yu, Zixin Yin, Deyu Zhou, Duomin Wang, Finn Wong, Baoyuan Wang

ICCV 2023 · Video › Video generation · approach: Talking face generation · 45 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
use diffusion prior instead of auto-regressive prior for non-lip motion (ablation)audio-driven talking head · VoxCeleb2SND: 7.65 -> 4.94Improved
use diffusion prior instead of auto-regressive prior (ablation)audio-driven talking head · VoxCeleb2FID: 24.31 -> 14.58Improved
use diffusion prior instead of auto-regressive prior (ablation)audio-driven talking head · VoxCeleb2 user studyNaturalness MOS: 2.53 -> 3.82Improved
add velocity loss to diffusion prior (ablation)diffusion prior · VoxCeleb2SND: 5.29 -> 4.94Improved
add mask editing during training (ablation)diffusion prior · VoxCeleb2SND: 8.76 -> 4.94Improved
add orthogonal loss between non-lip and audio-lip features (ablation)lip/non-lip disentanglement · VoxCeleb2SyncNet confidence Sc: 2.84 -> 7.48Improved
add memory banks for orthogonal loss (ablation)lip/non-lip disentanglement · VoxCeleb2pose distance Pd: 0.0293 -> 0.0019Improved
diffusion prior versus PC-AVS in image quality (FID)audio-driven talking head · VoxCeleb2FID: 14.36 (PC-AVS) vs 14.58No clear effect
diffusion prior versus Wav2Lip on SyncNet confidenceaudio-driven talking head · VoxCeleb2SyncNet confidence Sc: 9.23 (Wav2Lip) vs 7.57Hurt

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