A temporally aligned gloss-level conditioning method with local fusion improves lexical order and semantic accuracy in sign language generation, reporting state-of-the-art results on CSL-Daily and Phoenix-2014T.
The language of motion: Unifying verbal and non-verbal language of 3d human motion
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.CV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
SIGNER: Temporally Grounded Sign Language Generation via Time-Resolved Conditioning
A temporally aligned gloss-level conditioning method with local fusion improves lexical order and semantic accuracy in sign language generation, reporting state-of-the-art results on CSL-Daily and Phoenix-2014T.