- formatting
- images
- links
- math
- code
- blockquotes
•
•
•
•
•
Person-Specific Deepfakes with 3D Morphable Models
How 3D morphable models enable person-specific deepfake avatars — a technical overview of neural rendering and identity-preserving facial animation.
Wav2Lip: Generalized Lip Sync Models
An overview of Wav2Lip, a generalised deep learning model for lip sync that works across speakers and video domains.
The Many Forms of Deep, Audio-Driven Video Dubbing
A survey of deep learning methods for audio-driven video dubbing and talking head generation, including neural rendering and 3DMM-based approaches.