EUVIMEDEUVIMED entdecken
Crossref

FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery

Zhuo Zheng, Yanfei Zhong, Junjue Wang, Ailong Ma, Liangpei Zhang

IEEE Transactions on Pattern Analysis and Machine Intelligence · 01.11.2023 · 45 (11) · 13715-13729

DOI: 10.1109/tpami.2023.3296757 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, erschienen am 01.11.2023. Verfasst von Zhuo Zheng und 4 weiteren Autor:innen. Die Arbeit wurde bisher 98-mal zitiert (Crossref) und nennt 69 Literaturangaben. Gefördert von: National Natural Science Foundation of China, LIESMARS Special Research Funding.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Geospatial object segmentation, a fundamental Earth vision task, always suffers from scale variation, the larger intra-class variance of background, and foreground-background imbalance in high spatial resolution (HSR) remote sensing imagery. Generic semantic segmentation methods mainly focus on the scale variation in natural scenarios. However, the other two problems are insufficiently considered in large area Earth observation scenarios. In this paper, we propose a foreground-aware relation network (FarSeg++) from the perspectives of relation-based, optimization-based, and objectness-based foreground modeling, alleviating the above two problems. From the perspective of the relations, the foreground-scene relation module improves the discrimination of the foreground features via the foreground-correlated contexts associated with the object-scene relation. From the perspective of optimization, foreground-aware optimization is proposed to focus on foreground examples and hard examples of the background during training to achieve a balanced optimization. Besides, from the perspective of objectness, a foreground-aware decoder is proposed to improve the objectness representation, alleviating the objectness prediction problem that is the main bottleneck revealed by an empirical upper bound analysis. We also introduce a new large-scale high-resolution urban vehicle segmentation dataset to verify the effectiveness of the proposed method and push the development of objectness prediction further forward. The experimental results suggest that FarSeg++ is superior to the state-of-the-art generic semantic segmentation methods and can achieve a better trade-off between speed and accuracy.

Förderung

  • National Natural Science Foundation of China
  • LIESMARS Special Research Funding

Zitieren

APA

Zheng, Z., Zhong, Y., Wang, J., Ma, A., & Zhang, L. (2023). FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(11), 13715-13729. https://doi.org/10.1109/tpami.2023.3296757

Vancouver

Zheng Z, Zhong Y, Wang J, Ma A, Zhang L. FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023;45(11):13715-13729. doi:10.1109/tpami.2023.3296757

BibTeX
@article{Zheng2023,
  author = {Zheng, Zhuo and Zhong, Yanfei and Wang, Junjue and Ma, Ailong and Zhang, Liangpei},
  title = {FarSeg++: Foreground-Aware Relation Network for Geospatial Object Segmentation in High Spatial Resolution Remote Sensing Imagery},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2023},
  volume = {45},
  number = {11},
  pages = {13715--13729},
  doi = {10.1109/tpami.2023.3296757}
}
Kostenlos starten →