EUVIMEDEUVIMED entdecken
Crossref

RePIC-Seg: Reciprocal Prototype–Image Consistency-Driven Semantic Segmentation for Remote Sensing Images

Hao-Shuo Zhu, Yi-Han Xu, Yu-Li Liu, Wen Zhou

IEEE Geoscience and Remote Sensing Letters · 2026 · 23 · 1-5

DOI: 10.1109/lgrs.2026.3673074 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Geoscience and Remote Sensing Letters, 2026. Verfasst von Hao-Shuo Zhu und 3 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Jiangsu Graduate Research and Practice Innovation Program.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Remote sensing image semantic segmentation is crucial for applications such as land monitoring and urban governance, yet remains challenging due to high inter-class similarity, large intra-class variability, and complex high-resolution structures that often cause boundary leakage. To address these challenges, we propose a Reciprocal Prototype–Image Consistency-driven segmentation (RePIC-Seg) framework for remote sensing images. Specifically, we design a RePIC Interactor layer that establishes reciprocal consistency interactions between class prototypes and image features, enabling stable prototype updating and effective feature refinement, thereby alleviating prototype drift and cross-class leakage. Furthermore, a reciprocal margin loss (RML) is proposed to enlarge the margin between the ground-truth class and the most competitive negative class, further improving inter-class separability and boundary discrimination. Extensive experiments on the LoveDA, Potsdam, and Vaihingen datasets demonstrate that RePIC-Seg consistently achieves superior performance. In particular, it outperforms competing methods by 1.9% mIoU on LoveDA. The released code is available at https://github.com/Flyzhuhs/RePIC-Seg

Förderung

  • National Natural Science Foundation of China
  • Jiangsu Graduate Research and Practice Innovation Program

Zitieren

APA

Zhu, H., Xu, Y., Liu, Y., & Zhou, W. (2026). RePIC-Seg: Reciprocal Prototype–Image Consistency-Driven Semantic Segmentation for Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters, 23, 1-5. https://doi.org/10.1109/lgrs.2026.3673074

Vancouver

Zhu H, Xu Y, Liu Y, Zhou W. RePIC-Seg: Reciprocal Prototype–Image Consistency-Driven Semantic Segmentation for Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters. 2026;23:1-5. doi:10.1109/lgrs.2026.3673074

BibTeX
@article{Zhu2026,
  author = {Zhu, Hao-Shuo and Xu, Yi-Han and Liu, Yu-Li and Zhou, Wen},
  title = {RePIC-Seg: Reciprocal Prototype–Image Consistency-Driven Semantic Segmentation for Remote Sensing Images},
  journal = {IEEE Geoscience and Remote Sensing Letters},
  year = {2026},
  volume = {23},
  pages = {1--5},
  doi = {10.1109/lgrs.2026.3673074}
}
Kostenlos starten →