DiffuSAM synthesizes SAM2-compatible mask embeddings via a diffusion prior conditioned on prior slices to enable accurate prompt-free medical image segmentation under SF-UDA and few-shot settings.
Deep learning has achieved re- markable success in this task [1], but supervised approaches require large annotated datasets, which are costly and expert- dependent
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
-
DiffuSAM: Diffusion-Based Prompt-Free SAM2 for Few-Shot and Source-Free Medical Image Segmentation
DiffuSAM synthesizes SAM2-compatible mask embeddings via a diffusion prior conditioned on prior slices to enable accurate prompt-free medical image segmentation under SF-UDA and few-shot settings.