EUVIMEDEUVIMED entdecken
Crossref

Relative Depth-Guided Cross-Domain Semantic Segmentation for High-Resolution Satellite Remote Sensing Images

Yongqi Sun, Yujun Quan, Anzhu Yu, Yu Su, Xin Li, Chenguang Dai, Xuanguang Liu, Yanfei Zhong

IEEE Transactions on Geoscience and Remote Sensing · 2026 · 64 · 5626118-5626118

DOI: 10.1109/tgrs.2026.3701909 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2026. Verfasst von Yongqi Sun und 7 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 68 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Satellite remote sensing images exhibit significant cross-domain discrepancies due to variations in sensors, imaging conditions, and regional characteristics. These discrepancies degrade the performance of semantic segmentation models on unseen target domains. Unsupervised domain adaptation (UDA) mitigates this problem but requires target-domain images during training. Achieving robust cross-domain generalization (DG) with foundation models typically relies on large-scale architectures, incurring substantial computational overhead that limits practical deployment. This article introduces a novel perspective by leveraging zero-shot relative depth maps as domain-invariant geometric priors to guide cross-domain semantic segmentation of satellite remote sensing images. The proposed framework is compatible with most encoder–decoder architectures and makes three key contributions: 1) a unified depth-guided framework for optical remote sensing cross-domain segmentation; 2) a multiscale contrastive learning module that aligns optical and relative depth features, enhancing optical feature representation; and 3) a refinement strategy using shallow relative depth features to guide semantic mask prediction. Experiments across three cross-domain scenarios, including the LoveDA benchmark, a self-constructed circumpolar cross-seasonal dataset, and a cross-dataset generalization setting, demonstrate that relative depth serves as an effective spatial prior. Notably, the proposed method achieves comparable or superior accuracy to state-of-the-art UDA approaches that require target-domain images during training, while operating under the stricter zero-shot DG setting where no target data is accessible. The codes will be released at https://github.com/ArthasMil/DGCD.

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China

Zitieren

APA

Sun, Y., Quan, Y., Yu, A., Su, Y., Li, X., Dai, C., Liu, X., & Zhong, Y. (2026). Relative Depth-Guided Cross-Domain Semantic Segmentation for High-Resolution Satellite Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing, 64, 5626118-5626118. https://doi.org/10.1109/tgrs.2026.3701909

Vancouver

Sun Y, Quan Y, Yu A, Su Y, Li X, Dai C, et al. Relative Depth-Guided Cross-Domain Semantic Segmentation for High-Resolution Satellite Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing. 2026;64:5626118-5626118. doi:10.1109/tgrs.2026.3701909

BibTeX
@article{Sun2026,
  author = {Sun, Yongqi and Quan, Yujun and Yu, Anzhu and Su, Yu and Li, Xin and Dai, Chenguang and Liu, Xuanguang and Zhong, Yanfei},
  title = {Relative Depth-Guided Cross-Domain Semantic Segmentation for High-Resolution Satellite Remote Sensing Images},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2026},
  volume = {64},
  pages = {5626118--5626118},
  doi = {10.1109/tgrs.2026.3701909}
}
Kostenlos starten →