CATP prunes low-confidence tokens in COD Transformers and uses dual-path compensation to cut computation while preserving segmentation accuracy on boundary regions.
Advances in Neural Information Processing Systems37, 114765–114796 (2024) 5
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
CATP: Confidence-Aware Token Pruning for Camouflaged Object Detection
CATP prunes low-confidence tokens in COD Transformers and uses dual-path compensation to cut computation while preserving segmentation accuracy on boundary regions.