Reconstruction Guided Few-shot Network For Remote Sensing Image Classification

Mohit Jaiswal, Naman Jain, Shivani Pathak, Mainak Singha, Nikunja Bihari Kar +2 more
1/12/2026
cs.CV

Official code: stark0908/RGFS (1 star, last commit 2026-01-29).

Abstract

Few-shot remote sensing image classification is challenging due to limited labeled samples and high variability in land-cover types. We propose a reconstruction-guided few-shot network (RGFS-Net) that enhances generalization to unseen classes while preserving consistency for seen categories. Our method incorporates a masked image reconstruction task, where parts of the input are occluded and reconstructed to encourage semantically rich feature learning. This auxiliary task strengthens spatial understanding and improves class discrimination under low-data settings. We evaluated the efficacy of EuroSAT and PatternNet datasets under 1-shot and 5-shot protocols, our approach consistently outperforms existing baselines. The proposed method is simple, effective, and compatible with standard backbones, offering a robust solution for few-shot remote sensing classification. Codes are available at https://github.com/stark0908/RGFS.

View on arXivView PDF

Code Implementations(1)

stark0908/RGFSOfficial100%

All the codes from the rgfs experiments

10Python, Jupyter NotebookJul 30, 20258 months ago

Similar Papers

Adaptive Multi-Scale Correlation Meta-Network for Few-Shot Remote Sensing Image Classification

85% match2026

Generalized Few-Shot Meets Remote Sensing: Discovering Novel Classes in Land Cover Mapping via Hybrid Semantic Segmentation Framework

83% match1 repo432024

InfRS: Incremental Few-Shot Object Detection in Remote Sensing Images

83% match1 repo172024

Few-Shot Adaptation Benchmark for Remote Sensing Vision-Language Models

82% match1 repo192025

Generalized Few-Shot Semantic Segmentation in Remote Sensing: Challenge and Benchmark

82% match1 repo232024

Efficient Few-Shot Learning in Remote Sensing: Fusing Vision and Vision-Language Models

81% match2025

Cite this paper

@article{jaiswal2026reconstruction,
  title  = {Reconstruction Guided Few-shot Network For Remote Sensing Image Classification},
  author = {Mohit Jaiswal and Naman Jain and Shivani Pathak and Mainak Singha and Nikunja Bihari Kar and Ankit Jha and Biplab Banerjee},
  year   = {2026},
  eprint = {2601.07335},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2601.07335v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2601.07335.svg)](https://codeofpaper.com/paper/2601.07335)

Discussion