Domain Adaptation on Semantic Segmentation for Aerial Images

TL;DR

This paper introduces an unsupervised domain adaptation framework for aerial image semantic segmentation, addressing the challenge of domain shift due to environmental variability, and demonstrates improved accuracy over existing methods.

Contribution

The paper presents a novel unsupervised domain adaptation approach that learns soft label distribution differences and applies entropy minimization, specifically tailored for aerial semantic segmentation.

Findings

Improved segmentation accuracy on ISPRS dataset

Effective reduction of domain gap impacts

Outperforms state-of-the-art methods

Abstract

Semantic segmentation has achieved significant advances in recent years. While deep neural networks perform semantic segmentation well, their success rely on pixel level supervision which is expensive and time-consuming. Further, training using data from one domain may not generalize well to data from a new domain due to a domain gap between data distributions in the different domains. This domain gap is particularly evident in aerial images where visual appearance depends on the type of environment imaged, season, weather, and time of day when the environment is imaged. Subsequently, this distribution gap leads to severe accuracy loss when using a pretrained segmentation model to analyze new data with different characteristics. In this paper, we propose a novel unsupervised domain adaptation framework to address domain shift in the context of aerial semantic image segmentation. To this end, we solve the problem of domain shift by learn the soft label distribution difference between the source and target domains. Further, we also apply entropy minimization on the target domain to produce high-confident prediction rather than using high-confident prediction by pseudo-labeling. We demonstrate the effectiveness of our domain adaptation framework using the challenge image segmentation dataset of ISPRS, and show improvement over state-of-the-art methods in terms of various metrics.