A ViT layer-decomposition based self-disentanglement and re-composition method improves cross-domain few-shot segmentation, beating prior state-of-the-art by 1.92 (1-shot) and 1.88 (5-shot) average mIoU.
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
-
Self-Disentanglement and Re-Composition for Cross-Domain Few-Shot Segmentation
A ViT layer-decomposition based self-disentanglement and re-composition method improves cross-domain few-shot segmentation, beating prior state-of-the-art by 1.92 (1-shot) and 1.88 (5-shot) average mIoU.