A language-guided bootstrapped disentanglement framework reduces class and background entanglement in continual semantic segmentation, improving state-of-the-art on VOC and ADE20k.
Modeling the back- ground for incremental learning in semantic segmentation
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
background 1representative citing papers
citing papers explorer
-
Learning Yourself: Class-Incremental Semantic Segmentation with Language-Inspired Bootstrapped Disentanglement
A language-guided bootstrapped disentanglement framework reduces class and background entanglement in continual semantic segmentation, improving state-of-the-art on VOC and ADE20k.