DiffDecompose recovers foreground and background layers from alpha-composited images using in-context diffusion with position encoding cloning, trained and evaluated on a new six-task synthetic dataset.
Sequential modeling enables scalable learning for large vision models
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
-
DiffDecompose: Layer-Wise Decomposition of Alpha-Composited Images via Diffusion Transformers
DiffDecompose recovers foreground and background layers from alpha-composited images using in-context diffusion with position encoding cloning, trained and evaluated on a new six-task synthetic dataset.