CDG-MAE pretrains a masked autoencoder using diffusion-generated views of static images plus multi-anchor masking, improving cross-view correspondence on video label propagation without video 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
CONDITIONAL 1representative citing papers
citing papers explorer
-
CDG-MAE: Cross-view Masked Modeling using Diffusion Generated Views
CDG-MAE pretrains a masked autoencoder using diffusion-generated views of static images plus multi-anchor masking, improving cross-view correspondence on video label propagation without video data.