Talking Head Generation with Probabilistic Audio-to-Visual Diffusion Priors.
Zhentao YuZixin YinDeyu ZhouDuomin WangFinn WongBaoyuan WangPublished in: CoRR (2022)
Keyphrases
- visual information
- visual data
- cross modal
- learned from training data
- prior probabilities
- visual features
- multimedia
- visual cues
- audio visual
- probabilistic model
- diffusion process
- low level
- generation process
- anisotropic diffusion
- uncertain data
- signal processing
- bayesian networks
- real time
- multimodal information
- generalized em algorithm
- multi modal
- video sequences
- visual perception
- eye tracking
- visual field
- audio files
- video indexing and retrieval
- social networks