MergeTalk: Audio-Driven Talking Head Generation From Single Image With Feature Merge

Keyphrases

• Audio-driven Talking Head Generation 100%
• Lip Synchronization 66%
• Pose Parameters 33%
• Lip Movement 33%
• Merging Network 33%
• Generation Network 33%
• Facial Distortion 33%
• Facial Artifacts 33%
• Synchronization Problem 33%
• 3D Morphable Model (3DMM) 33%
• 3D Motion Parameters 33%
• Talking Head 33%
• Head Quality 33%
• Parameter Generation 33%
• Facial Texture 33%
• Identity Preservation 33%

Computer Science

• Pose Parameter 50%