X-Actor generates long, emotionally expressive talking-head videos from a single portrait and audio by autoregressively diffusing compact facial motion latents before video synthesis.
We use six metrics for quantitative evaluation, with ground-truth comparisons available in the self- driven setting
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
X-Actor: Emotional and Expressive Long-Range Portrait Acting from Audio
X-Actor generates long, emotionally expressive talking-head videos from a single portrait and audio by autoregressively diffusing compact facial motion latents before video synthesis.