An object-centric video model learns temporally consistent slots that condition a frozen diffusion decoder, enabling reconstruction and compositional editing of real-world videos.
Stretchbev: Stretching future instance prediction spatially and temporally
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
-
Compositional Video Synthesis by Temporal Object-Centric Learning
An object-centric video model learns temporally consistent slots that condition a frozen diffusion decoder, enabling reconstruction and compositional editing of real-world videos.