Dual Contrastive Network for Few-Shot Remote Sensing Image Scene Classification

2026-03-24

AI Summary

This work addresses the challenges of small inter-class variation and large intra-class variation in few-shot scene classification of remote sensing images by proposing a transfer learning–based Dual Contrastive Network (DCN). The method introduces two novel contrastive learning branches: a context-guided branch that employs a Condenser Network to capture global contextual information and enhance inter-class separability, and a detail-guided assistant branch that leverages a Smelter Network to focus on local details and improve intra-class invariance. Both branches are integrated within a unified framework and optimized through supervised contrastive learning. Extensive experiments on four public remote sensing datasets demonstrate that the proposed approach significantly outperforms existing methods under few-shot settings, confirming its strong discriminative power and robustness.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Scene Analysis & Understanding
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Few-shot remote sensing image scene classification (FS-RSISC) aims at classifying remote sensing images with only a few labeled samples. The main challenges lie in small inter-class variances and large intra-class variances, which are the inherent property of remote sensing images. To address these challenges, we propose a transfer-based Dual Contrastive Network (DCN), which incorporates two auxiliary supervised contrastive learning branches during the training process. Specifically, one is a Context-guided Contrastive Learning (CCL) branch and the other is a Detail-guided Contrastive Learning (DCL) branch, which focus on inter-class discriminability and intra-class invariance, respectively. In the CCL branch, we first devise a Condenser Network to capture context features, and then leverage a supervised contrastive learning on top of the obtained context features to facilitate the model to learn more discriminative features. In the DCL branch, a Smelter Network is designed to highlight the significant local detail information. And then we construct a supervised contrastive learning based on the detail feature maps to fully exploit the spatial information in each map, enabling the model to concentrate on invariant detail features. Extensive experiments on four public benchmark remote sensing datasets demonstrate the competitive performance of our proposed DCN.

Problem

• Few-shot learning
• Remote sensing image classification
• Inter-class variance
• Intra-class variance
• Scene classification

Innovation

• Dual Contrastive Learning
• Few-Shot Learning
• Remote Sensing Image Classification
• Context-Guided Feature Learning
• Detail-Guided Feature Learning