Cross-Domain Few-Shot Learning for Hyperspectral Image Classification Based on Mixup Foundation Model

2026-01-30

AI Summary

This work addresses the challenges of cross-domain few-shot classification in hyperspectral imagery—namely, severe data scarcity, reliance on unrealistic data augmentation, and susceptibility to overfitting—by introducing, for the first time, a remote sensing foundation model into this task. The authors propose an efficient adaptation framework that freezes the backbone network and incorporates a condensed projection mechanism for rapid downstream adaptation. To mitigate inter-domain distribution shifts, they employ Mixup-based domain adaptation, while label smoothing is utilized to suppress noise from pseudo-labels. Evaluated across multiple benchmarks, the proposed method substantially outperforms existing approaches, achieving performance gains of up to 14%. The code has been made publicly available to support reproducible research.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Remote Sensing / Geospatial AI
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web applications in cross-disciplinary domains and verticals such as mixed reality, smart cities, and digital health

Abstract

Although cross-domain few-shot learning (CDFSL) for hyper-spectral image (HSI) classification has attracted significant research interest, existing works often rely on an unrealistic data augmentation procedure in the form of external noise to enlarge the sample size, thus greatly simplifying the issue of data scarcity. They involve a large number of parameters for model updates, being prone to the overfitting problem. To the best of our knowledge, none has explored the strength of the foundation model, having strong generalization power to be quickly adapted to downstream tasks. This paper proposes the MIxup FOundation MOdel (MIFOMO) for CDFSL of HSI classifications. MIFOMO is built upon the concept of a remote sensing (RS) foundation model, pre-trained across a large scale of RS problems, thus featuring generalizable features. The notion of coalescent projection (CP) is introduced to quickly adapt the foundation model to downstream tasks while freezing the backbone network. The concept of mixup domain adaptation (MDM) is proposed to address the extreme domain discrepancy problem. Last but not least, the label smoothing concept is implemented to cope with noisy pseudo-label problems. Our rigorous experiments demonstrate the advantage of MIFOMO, where it beats prior arts with up to 14% margin. The source code of MIFOMO is open-sourced in https://github.com/Naeem- Paeedeh/MIFOMO for reproducibility and convenient further study.

Problem

• Cross-Domain Few-Shot Learning
• Hyperspectral Image Classification
• Foundation Model
• Data Scarcity
• Domain Discrepancy

Innovation

• foundation model
• cross-domain few-shot learning
• mixup domain adaptation
• coalescent projection
• hyperspectral image classification