Fine-tuning a deep dense matching network (RoMa) on a newly built airborne SAR image dataset improves stereo radargrammetry accuracy and coverage compared to POC-based and camera-only-trained baselines.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Stereo Radargrammetry Using Deep Learning from Airborne SAR Images
Fine-tuning a deep dense matching network (RoMa) on a newly built airborne SAR image dataset improves stereo radargrammetry accuracy and coverage compared to POC-based and camera-only-trained baselines.