A unified 3D talking-head model that switches between speaker and listener roles improves naturalness of dyadic conversations on a new 50-hour multi-round dataset.
High-fidelity fa- cial avatar reconstruction from monocular video with gen- erative priors
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
DualTalk: Dual-Speaker Interaction for 3D Talking Head Conversations
A unified 3D talking-head model that switches between speaker and listener roles improves naturalness of dyadic conversations on a new 50-hour multi-round dataset.