Synthetic action videos generated by pose-transferring real clips onto novel 3D avatars improve action recognition accuracy when added to real training data.
Is space-time attention all you need for video understanding? InICML, page 4, 2021
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
-
Synthetic Human Action Video Data Generation with Pose Transfer
Synthetic action videos generated by pose-transferring real clips onto novel 3D avatars improve action recognition accuracy when added to real training data.