LAST: Utilizing Synthetic Image Style Transfer to Tackle Domain Shift in Aerial Image Segmentation

Yubo Wang, Ruijia Wen, Hiroyuki Ishii, Jun Ohya

2025

Abstract

Recent deep learning models often struggle with performance degradation due to domain shifts. Addressing domain adaptation in aerial image segmentation is challenging due to the limited availability of training data. To tackle this, we utilized the Unreal Engine to construct a synthetic dataset featuring images captured under diverse conditions such as fog, snow, and nighttime settings. We then proposed a latent space style transfer model that generates alternate domain versions based on the real aerial dataset. This approach eliminates the need for additional annotations on shifted domain data. We benchmarked nine different state-of-the-art segmentation methods on the ISPRS Vaihingen, Potsdam datasets, and their shifted foggy domains. Extensive experiments reveal that domain shift leads to significant performance drops, with an average decrease of - 3.46% mIoU on Vaihingen and -5.22% mIoU on Potsdam. Finally, we adapted the model to perform well in the shifted domain, achieving improvements of +2.97% mIoU on Vaihingen and +3.97% mIoU on Potsdam, while maintaining its effectiveness in the original domain.

Paper Citation

in Harvard Style

Wang Y., Wen R., Ishii H. and Ohya J. (2025). LAST: Utilizing Synthetic Image Style Transfer to Tackle Domain Shift in Aerial Image Segmentation. In Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM; ISBN 978-989-758-730-6, SciTePress, pages 32-42. DOI: 10.5220/0013145000003905

in Bibtex Style

@conference{icpram25,
author={Yubo Wang and Ruijia Wen and Hiroyuki Ishii and Jun Ohya},
title={LAST: Utilizing Synthetic Image Style Transfer to Tackle Domain Shift in Aerial Image Segmentation},
booktitle={Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM},
year={2025},
pages={32-42},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013145000003905},
isbn={978-989-758-730-6},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 14th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM
TI - LAST: Utilizing Synthetic Image Style Transfer to Tackle Domain Shift in Aerial Image Segmentation
SN - 978-989-758-730-6
AU - Wang Y.
AU - Wen R.
AU - Ishii H.
AU - Ohya J.
PY - 2025
SP - 32
EP - 42
DO - 10.5220/0013145000003905
PB - SciTePress