A pretrained 14B image-to-video diffusion model is adapted into a two-step autoregressive audio-driven avatar generator that can stream indefinitely in real time.
Video generation models as world simulators, 2024
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.SD 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
TalkingMachines: Real-Time Audio-Driven FaceTime-Style Video via Autoregressive Diffusion Models
A pretrained 14B image-to-video diffusion model is adapted into a two-step autoregressive audio-driven avatar generator that can stream indefinitely in real time.