A SAR-trained YOLOv11 detector supplies box prompts to SAM2 for zero-shot ship instance segmentation, reaching 0.637 mean IoU (89% of supervised baseline) and 89.2% detection rate on SSDD.
LS-SSDD-v1.0: A Deep Learn- ing Dataset Dedicated to Small Ship Detection from Large-Scale Sentinel-1 SAR Images
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
-
Prompting Foundation Models for Zero-Shot Ship Instance Segmentation in SAR Imagery
A SAR-trained YOLOv11 detector supplies box prompts to SAM2 for zero-shot ship instance segmentation, reaching 0.637 mean IoU (89% of supervised baseline) and 89.2% detection rate on SSDD.