EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Integrated Framework for Unsupervised Building Segmentation with Segment Anything Model-Based Pseudo-Labeling and Weakly Supervised Learning

Jiyong Kim, Yongil Kim

Remote Sensing · 30.01.2024 · 16 (3) · 526

DOI: 10.3390/rs16030526 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 30.01.2024. Verfasst von Jiyong Kim und 1 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Research Foundation of Korea (NRF), Korea Agency for Infrastructure Technology Advancement (KAIA), BK21 FOUR research program of the National Research Foundation of Korea.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The Segment Anything Model (SAM) has had a profound impact on deep learning applications in remote sensing. SAM, which serves as a prompt-based foundation model for segmentation, exhibits a remarkable capability to “segment anything,” including building objects on satellite or airborne images. To facilitate building segmentation without inducing supplementary prompts or labels, we applied a sequential approach of generating pseudo-labels and incorporating an edge-driven model. We first segmented the entire scene by SAM and masked out unwanted objects to generate pseudo-labels. Subsequently, we employed an edge-driven model designed to enhance the pseudo-label by using edge information to reconstruct the imperfect building features. Our model simultaneously utilizes spectral features from SAM-oriented building pseudo-labels and edge features from resultant images from the Canny edge detector and, thus, when combined with conditional random fields (CRFs), shows capability to extract and learn building features from imperfect pseudo-labels. By integrating the SAM-based pseudo-label with our edge-driven model, we establish an unsupervised framework for building segmentation that operates without explicit labels. Our model excels in extracting buildings compared with other state-of-the-art unsupervised segmentation models and even outperforms supervised models when trained in a fully supervised manner. This achievement demonstrates the potential of our model to address the lack of datasets in various remote sensing domains for building segmentation.

Förderung

  • National Research Foundation of Korea (NRF)
  • Korea Agency for Infrastructure Technology Advancement (KAIA)
  • BK21 FOUR research program of the National Research Foundation of Korea

Zitieren

APA

Kim, J., & Kim, Y. (2024). Integrated Framework for Unsupervised Building Segmentation with Segment Anything Model-Based Pseudo-Labeling and Weakly Supervised Learning. Remote Sensing, 16(3), 526. https://doi.org/10.3390/rs16030526

Vancouver

Kim J, Kim Y. Integrated Framework for Unsupervised Building Segmentation with Segment Anything Model-Based Pseudo-Labeling and Weakly Supervised Learning. Remote Sensing. 2024;16(3):526. doi:10.3390/rs16030526

BibTeX
@article{Kim2024,
  author = {Kim, Jiyong and Kim, Yongil},
  title = {Integrated Framework for Unsupervised Building Segmentation with Segment Anything Model-Based Pseudo-Labeling and Weakly Supervised Learning},
  journal = {Remote Sensing},
  year = {2024},
  volume = {16},
  number = {3},
  pages = {526},
  doi = {10.3390/rs16030526}
}
Kostenlos starten →