EUVIMEDEUVIMED entdecken
Crossref

Incorporating Multiscale Context and Task-Consistent Focal Loss into Oriented Object Detection

Xiaoliang Qian, Qingqing Jian, Wei Wang, Xiwen Yao, Gong Cheng

IEEE Transactions on Geoscience and Remote Sensing · 2025 · 63 · 1-11

DOI: 10.1109/tgrs.2025.3580937 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Geoscience and Remote Sensing, 2025. Verfasst von Xiaoliang Qian und 4 weiteren Autor:innen. Die Arbeit wurde bisher 57-mal zitiert (Crossref) und nennt 56 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Henan Province, Key Research Project of Henan Province Universities.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Oriented object detection (OOD) in remote sensing images (RSIs) aims to precisely localize and identify objects with arbitrary orientations. Two-stage OOD methods attract lots of interest due to their superior accuracy; however, they still face two major problems. First of all, the misclassification problem frequently occurs because the majority of classification strategies solely rely on the features of proposals. Second, most loss functions cannot simultaneously concentrate on hard samples and boost the consistency between identification and localization, which restricts the further improvement of OOD models. To address the first problem, the multiscale context (MSC) is incorporated into a two-stage OOD model in this article. Specifically, N contextual branches are added to predict the class confidence score (CCS) of each proposal and its N enlarged proposals which contain the MSC, and the final CCS of each proposal is determined by the mean value of above $N+1$ CCSs. To tackle the second problem, a task-consistent focal (TF) loss is proposed. The TF loss employs the difficulty of localization as the weight of classification loss, and the difficulty of identification is used as the weight of regression loss. Concentrating on hard samples and synchronous optimization of classification and regression can be achieved by minimizing the TF loss. The ablation studies show the validity of MSC, TF, and their combination. The comparison with popular OOD models demonstrates the superior performance of our model on the DOTA and DIOR-R datasets. The source code can be obtained from https://github.com/qxlzengli/MSC-TF

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Henan Province
  • Key Research Project of Henan Province Universities

Zitieren

APA

Qian, X., Jian, Q., Wang, W., Yao, X., & Cheng, G. (2025). Incorporating Multiscale Context and Task-Consistent Focal Loss into Oriented Object Detection. IEEE Transactions on Geoscience and Remote Sensing, 63, 1-11. https://doi.org/10.1109/tgrs.2025.3580937

Vancouver

Qian X, Jian Q, Wang W, Yao X, Cheng G. Incorporating Multiscale Context and Task-Consistent Focal Loss into Oriented Object Detection. IEEE Transactions on Geoscience and Remote Sensing. 2025;63:1-11. doi:10.1109/tgrs.2025.3580937

BibTeX
@article{Qian2025,
  author = {Qian, Xiaoliang and Jian, Qingqing and Wang, Wei and Yao, Xiwen and Cheng, Gong},
  title = {Incorporating Multiscale Context and Task-Consistent Focal Loss into Oriented Object Detection},
  journal = {IEEE Transactions on Geoscience and Remote Sensing},
  year = {2025},
  volume = {63},
  pages = {1--11},
  doi = {10.1109/tgrs.2025.3580937}
}
Kostenlos starten →