Phenaki generates arbitrary-length videos from sequences of text prompts by tokenizing videos with causal temporal attention and generating tokens with a text-conditioned masked transformer, trained jointly on images and videos.
N ¨Uwa: Visual synthesis pre-training for neural visual world creation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2022 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Phenaki: Variable Length Video Generation From Open Domain Textual Description
Phenaki generates arbitrary-length videos from sequences of text prompts by tokenizing videos with causal temporal attention and generating tokens with a text-conditioned masked transformer, trained jointly on images and videos.