A spectral vision transformer achieves equitable or superior performance with fewer parameters than standard ViTs, CNNs, and other models by using spectral projections for tokenization in limited-data medical imaging.
Wave-vit: Unifying wavelet and transformers for visual representation learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2roles
background 1polarities
background 1representative citing papers
FSS-Net uses wavelet attention and adaptive edge fusion modules to reach 96.46% Dice score on carotid ultrasound segmentation with reported robustness to low SNR.
citing papers explorer
-
Spectral Vision Transformer for Efficient Tokenization with Limited Data
A spectral vision transformer achieves equitable or superior performance with fewer parameters than standard ViTs, CNNs, and other models by using spectral projections for tokenization in limited-data medical imaging.
-
FSS-Net: Frequency-Spatial Synergy Network with Wavelet Attention for Carotid Artery Ultrasound Segmentation
FSS-Net uses wavelet attention and adaptive edge fusion modules to reach 96.46% Dice score on carotid ultrasound segmentation with reported robustness to low SNR.