LFINet achieves state-of-the-art F1 of 92.54% and IoU of 86.12% on real agricultural trajectory data by decoupling and interacting low- and high-frequency features to preserve topology.
Semantic segmen- tation for remote sensing images based on an AD-HRNet model
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
-
Laplacian Frequency Interaction Network for Rural Thematic Road Extraction
LFINet achieves state-of-the-art F1 of 92.54% and IoU of 86.12% on real agricultural trajectory data by decoupling and interacting low- and high-frequency features to preserve topology.