SGTC uses CLIP semantic guidance and three-view co-training to segment medical volumes from just three annotated slices per volume, with gains over baselines mainly driven by the extra label.
Title resolution pending
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
-
SGTC: Semantic-Guided Triplet Co-training for Sparsely Annotated Semi-Supervised Medical Image Segmentation
SGTC uses CLIP semantic guidance and three-view co-training to segment medical volumes from just three annotated slices per volume, with gains over baselines mainly driven by the extra label.