Frugal Learning of Virtual Exemplars for Label-Efficient Satellite Image Change Detection

TL;DR

This paper introduces a novel active learning framework for satellite image change detection that uses virtual exemplars to efficiently identify the most informative images, improving detection accuracy with fewer labeled samples.

Contribution

It proposes a new display model that selects diverse virtual exemplars to enhance change detection, advancing interactive satellite image analysis methods.

Findings

Outperforms existing methods in accuracy and efficiency

Effective in reducing the number of queries needed

Demonstrates robustness on challenging satellite datasets

Abstract

In this paper, we devise a novel interactive satellite image change detection algorithm based on active learning. The proposed framework is iterative and relies on a question and answer model which asks the oracle (user) questions about the most informative display (subset of critical images), and according to the user's responses, updates change detections. The contribution of our framework resides in a novel display model which selects the most representative and diverse virtual exemplars that adversely challenge the learned change detection functions, thereby leading to highly discriminating functions in the subsequent iterations of active learning. Extensive experiments, conducted on the challenging task of interactive satellite image change detection, show the superiority of the proposed virtual display model against the related work.