A training-free multi-signal pipeline using RAFT, DINOv2, ViBe, and SAM2 achieves 0.425041 mIoU and 0.457206 mDice on the DOST turbulence segmentation challenge.
Title resolution pending
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
-
Turbulence-Robust Dynamic Object Segmentation with Multi-Signal Priors and SAM2 Refinement
A training-free multi-signal pipeline using RAFT, DINOv2, ViBe, and SAM2 achieves 0.425041 mIoU and 0.457206 mDice on the DOST turbulence segmentation challenge.