SpectralEarth: Training Hyperspectral Foundation Models at Scale

Abstract

Foundation models have triggered a paradigm shift in computer vision and are increasingly being adopted in remote sensing, particularly for multispectral imagery. Yet, their potential in hyperspectral imaging (HSI) remains untapped due to the absence of comprehensive and globally representative hyperspectral datasets. To close this gap, we introduce SpectralEarth, a large-scale multitemporal dataset designed to pretrain hyperspectral foundation models leveraging data from the environmental mapping and analysis program (EnMAP). SpectralEarth comprises 538 974 image patches covering 415 153 unique locations from 11 636 globally distributed EnMAP scenes spanning two years of archive. In addition, 17.5% of these locations include multiple timestamps, enabling multitemporal HSI analysis. Utilizing state-of-the-art self-supervised learning algorithms, we pretrain a series of foundation models on SpectralEarth, integrating a spectral adapter into classical vision backbones to accommodate the unique characteristics of HSI. In tandem, we construct nine downstream datasets for land-cover, crop-type mapping, and tree-species classification, providing benchmarks for model evaluation. Experimental results support the versatility of our models and their generalizability across different tasks and sensors. We also highlight computational efficiency during model fine-tuning.

References in corpus (11)

• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Graph Convolutional Networks for Hyperspectral Image Classification
• DINOv2: Learning Robust Visual Features without Supervision
• SpectralGPT: Spectral Remote Sensing Foundation Model
• Deep Learning for Classification of Hyperspectral Data: A Comparative Review
• BigEarthNet-MM: A Large Scale Multi-Modal Multi-Label Benchmark Archive for Remote Sensing Image Classification and Retrieval
• Surgical Fine-Tuning Improves Adaptation to Distribution Shifts
• HySpecNet-11k: A Large-Scale Hyperspectral Dataset for Benchmarking Learning-Based Hyperspectral Image Compression Methods
• Prithvi-EO-2.0: A Versatile Multi-Temporal Foundation Model for Earth Observation Applications
• Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observation
• When are Foundation Models Effective? Understanding the Suitability for Pixel-Level Classification Using Multispectral Imagery