Mirage generates photorealistic A-roll videos of people speaking directly from audio, using only joint self-attention over audio, text, and video tokens.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Seeing Voices: Generating A-Roll Video from Audio with Mirage
Mirage generates photorealistic A-roll videos of people speaking directly from audio, using only joint self-attention over audio, text, and video tokens.