VeloxSeg pairs multi-scale window attention with JL-lemma-motivated grouped convolutions and Gram-matrix distillation to achieve efficient 3D medical segmentation at 1.66M parameters.
Our model improves the Dice by 1.72% compared to the state-of-the-art SuperLightNet
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
-
Johnson-Lindenstrauss Lemma Guided Network for Efficient 3D Medical Segmentation
VeloxSeg pairs multi-scale window attention with JL-lemma-motivated grouped convolutions and Gram-matrix distillation to achieve efficient 3D medical segmentation at 1.66M parameters.