Pretrained VAE latents have far lower transmission–reflection coherence than pixels, so flow matching on FLUX plus cycle and contrastive losses can separate both layers in one pass.
In: CVPR (2014)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
PRISM: Latent Composition Consistency for Single-Image Reflection Removal
Pretrained VAE latents have far lower transmission–reflection coherence than pixels, so flow matching on FLUX plus cycle and contrastive losses can separate both layers in one pass.