Active Learning for Improved Semi-Supervised Semantic Segmentation in Satellite Images

TL;DR

This paper introduces an active learning approach to enhance semi-supervised semantic segmentation of satellite images, significantly reducing the need for labeled data while maintaining high accuracy.

Contribution

It proposes an active learning sampling strategy to select diverse labeled data, improving semi-supervised segmentation performance on satellite imagery datasets.

Findings

27% improvement in mIoU with 2% labeled data

Effective selection of representative training samples

Demonstrated on UC Merced and DeepGlobe datasets

Abstract

Remote sensing data is crucial for applications ranging from monitoring forest fires and deforestation to tracking urbanization. Most of these tasks require dense pixel-level annotations for the model to parse visual information from limited labeled data available for these satellite images. Due to the dearth of high-quality labeled training data in this domain, there is a need to focus on semi-supervised techniques. These techniques generate pseudo-labels from a small set of labeled examples which are used to augment the labeled training set. This makes it necessary to have a highly representative and diverse labeled training set. Therefore, we propose to use an active learning-based sampling strategy to select a highly representative set of labeled training data. We demonstrate our proposed method's effectiveness on two existing semantic segmentation datasets containing satellite images: UC Merced Land Use Classification Dataset and DeepGlobe Land Cover Classification Dataset. We report a 27% improvement in mIoU with as little as 2% labeled data using active learning sampling strategies over randomly sampling the small set of labeled training data.