EAMM: One-Shot Emotional Talking Face via Audio-based Emotion-Aware Motion Model
Event Type
Technical Paper
Interest Areas
Research & Education
Presentation Types
In Person
Registration Categories
Full Conference Supporter
Full Conference
Exhibitor Additional Full Conference
Exhibitor Full Conference
This session WILL NOT be recorded.
TimeThursday, 11 August 20224:47pm - 4:52pm PDT
LocationEast Building, Ballroom A/B
DescriptionWe present Emotion-Aware Motion Model (EAMM) to generate one-shot emotional talking faces by leveraging an emotion source video. The key is to model the local emotion deformations by using the first-order motion representations. Experiments show that our method can generate satisfactory talking face results on arbitrary subjects with realistic emotions.