HyperFM: An Efficient Hyperspectral Foundation Model with Spectral Grouping

2026-04-22

AI Summary

This work addresses the challenges posed by NASA’s PACE mission hyperspectral data—namely, its large scale, scarce annotations, strong spectral continuity, and cross-sensor inconsistencies—which hinder the generalization and efficiency of existing foundation models. To overcome these limitations, we propose a parameter-efficient hyperspectral foundation model that introduces a novel spectral grouping attention mechanism coupled with hybrid parameter decomposition to effectively capture intra- and inter-group spectral-spatial dependencies. This design enables unified modeling of both cloudy and cloud-free scenes while substantially reducing computational costs. By integrating hyperspectral self-supervised pretraining with cross-sensor alignment, our model outperforms current hyperspectral foundation models and task-specific state-of-the-art methods across four atmospheric cloud property retrieval tasks. We also release HyperFM250K, a PACE-derived dataset comprising 250,000 samples.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Computer Vision: Other Foundations of Computer Vision
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

The NASA PACE mission provides unprecedented hyperspectral observations of ocean color, aerosols, and clouds, offering new insights into how these components interact and influence Earth's climate and air quality. Its Ocean Color Instrument measures light across hundreds of finely spaced wavelength bands, enabling detailed characterization of features such as phytoplankton composition, aerosol properties, and cloud microphysics. However, hyperspectral data of this scale is large, complex, and difficult to label, requiring specialized processing and analysis techniques. Existing foundation models, which have transformed computer vision and natural language processing, are generally trained on standard RGB imagery and therefore struggle to interpret the continuous spectral signatures captured by PACE. While recent advances have introduced hyperspectral foundation models, they are typically trained on cloud-free observations and often remain limited to single-sensor datasets due to spectral inconsistencies across instruments. Moreover, existing models tend to be parameter-heavy and computationally expensive, limiting scalability and adoption in operational settings. To address these challenges, we introduce HyperFM, a parameter-efficient hyperspectral foundation model that leverages intra-group and inter-group spectral attention along with hybrid parameter decomposition to better capture spectral spatial relationships while reducing computational cost. HyperFM demonstrates consistent performance improvements over existing hyperspectral foundation models and task-specific state-of-the-art methods across four benchmark downstream atmospheric cloud property retrieval tasks. To support further research, we additionally release HyperFM250K, a large-scale hyperspectral dataset from the PACE mission that includes both clear and cloudy scenes.

Problem

• hyperspectral
• foundation model
• spectral inconsistency
• computational efficiency
• cloudy scenes

Innovation

• hyperspectral foundation model
• spectral grouping
• parameter-efficient modeling
• spectral attention
• PACE mission