TL;DR
AgriFM is a novel multi-source, multi-scale transformer-based foundation model that effectively captures spatiotemporal patterns for accurate crop mapping from satellite data.
Contribution
It introduces a hierarchical spatiotemporal transformer architecture tailored for agricultural mapping, integrating multi-source satellite data and a dynamic fusion decoder.
Findings
Outperforms existing models in crop mapping accuracy
Efficient processing of long-term satellite time-series
Demonstrates versatility across diverse agricultural tasks
Abstract
Accurate crop mapping fundamentally relies on modeling multi-scale spatiotemporal patterns, where spatial scales range from individual field textures to landscape-level context, and temporal scales capture both short-term phenological transitions and full growing-season dynamics. Transformer-based remote sensing foundation models (RSFMs) offer promising potential for crop mapping due to their innate ability for unified spatiotemporal processing. However, current RSFMs remain suboptimal for crop mapping: they either employ fixed spatiotemporal windows that ignore the multi-scale nature of crop systems or completely disregard temporal information by focusing solely on spatial patterns. To bridge these gaps, we present AgriFM, a multi-source remote sensing foundation model specifically designed for agricultural crop mapping. Our approach begins by establishing the necessity of simultaneous…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
MethodsAttention Is All You Need · Linear Layer · Layer Normalization · Byte Pair Encoding · Stochastic Depth · Residual Connection · Dense Connections · Swin Transformer · Softmax · Position-Wise Feed-Forward Layer
