VpbSD uses a vessel-pattern codebook trained on unlabeled microscopy data to distill knowledge from a large teacher into a 0.12M-parameter student, reaching DSC 0.852 on VesSep2020.
Our strategies effectively leveraged feature learning from difficult- to-annotate microscopic images, resulting in a student model capable of fast and efficient segmentation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
VPBSD:Vessel-Pattern-Based Semi-Supervised Distillation for Efficient 3D Microscopic Cerebrovascular Segmentation
VpbSD uses a vessel-pattern codebook trained on unlabeled microscopy data to distill knowledge from a large teacher into a 0.12M-parameter student, reaching DSC 0.852 on VesSep2020.