KAN-FIF: Spline-Parameterized Lightweight Physics-based Tropical Cyclone Estimation on Meteorological Satellite

Dimensions overall score 8.0

GitHub Code Pulse

No public code linked for this paper yet.

• The KAN-FIF framework achieves a 94.8% reduction in parameters compared to the baseline model Phy-CoCo.result 95%
• The KAN-FIF framework provides 68.7% faster inference per sample compared to the baseline model Phy-CoCo.result 95%
• The KAN-FIF framework maintains superior accuracy with 32.5% lower MAE for Maximum Sustained Wind prediction compared to the baseline.result 95%
• The KAN-FIF framework achieved 14.41ms per-sample inference latency when deployed on the Qingyun-1000 development board with FY-4 series meteorological satellite processor.technical 90%
• Current physics-guided models suffer from linear feature interactions that fail to capture high-order polynomial relationships between TC attributes.limitation 85%
• The KAN-FIF framework demonstrates promising feasibility for operational TC monitoring and extending deployability to edge-device AI applications.market 80%
• Real-world satellite integration may face unforeseen compatibility issues.limitation 75%
• The KAN-FIF framework integrates MLP and CNN layers with spline-parameterized KAN layers.method 90%

Competitive landscape

Competitor map is still being generated for this paper. Enable generation or check back soon.

Startup potential card