DAE-Talker: High Fidelity Speech-Driven Talking Face Generation with Diffusion Autoencoder.
Chenpeng DuQi ChenTianyu HeXu TanXie ChenKai YuSheng ZhaoJiang BianPublished in: CoRR (2023)
Keyphrases
- high fidelity
- real time
- recognition engine
- medical image compression
- speech recognition
- real environment
- human faces
- high quality
- anisotropic diffusion
- face images
- audio visual
- speech signal
- video conferencing
- ground truth
- facial expressions
- high resolution
- automatic speech recognition
- multiresolution
- image analysis
- video sequences
- image processing