A motion-first framework builds a signer-independent lexicon of discrete sign motions and learns smooth transitions, enabling video generation of arbitrary signers from text without signer-specific sentence data.
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
-
Motion is the Choreographer: Learning Latent Pose Dynamics for Seamless Sign Language Generation
A motion-first framework builds a signer-independent lexicon of discrete sign motions and learns smooth transitions, enabling video generation of arbitrary signers from text without signer-specific sentence data.