Cite
Notes
Only stored in your browser.
Attribution
FLOAT: Generative Motion Latent Flow Matching for Audio-driven Talking Portrait
ICCV 2025
from 1 papers
Dongchan Min
Taekyung Ki