EUVIMEDEUVIMED entdecken
Crossref

Continual NeRF-Based 3D ISAR Imaging With Multilevel Distillation

Jianqiang Xu, Yulai Cong, Junyuan Deng, Fei Zeng, Mingcheng Dai

IEEE Signal Processing Letters · 2025 · 32 · 3495-3499

DOI: 10.1109/lsp.2025.3601535 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Signal Processing Letters, 2025. Verfasst von Jianqiang Xu und 4 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 36 Literaturangaben. Gefördert von: Natural Science Foundation of Guangdong Province, Guangdong Provincial Pearl River Talents Program, Guangdong Science and Technology Departmentunder.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In 3D inverse synthetic aperture radar (ISAR) imaging of space targets, radar neural radiance fields (i.e., ISAR-NeRF) is an important research direction. However, a significant yet unexplored problem is its deployment in continual learning scenarios, where gradual 3D imaging is expected since target ISAR images often emerge sequentially. To address this issue, this letter proposes a new continual 3D ISAR imaging method, named CL-ISAR-NeRF. Specifically, CL-ISAR-NeRF leverages a multilevel distillation mechanism to simultaneously replay pixel, field, and feature-levels information, to alleviate the forgetting of previously learned knowledge. In addition, an efficient memory selection strategy is designed to enrich the diversity of line-of-sight (LOS) when selecting replayed data, which improves imaging performance and further enhances the stability and plasticity of the method. In order to evaluate the proposed method in continual learning settings, we design a realistic simulation scenario in which the trajectories of space targets are calculated by the Simplified General Perturbations-4 (SGP4) model. The comparative experiments with classic continual learning methods demonstrate the superior performance and robustness of CL-ISAR-NeRF.

Förderung

  • Natural Science Foundation of Guangdong Province
  • Guangdong Provincial Pearl River Talents Program
  • Guangdong Science and Technology Departmentunder

Zitieren

APA

Xu, J., Cong, Y., Deng, J., Zeng, F., & Dai, M. (2025). Continual NeRF-Based 3D ISAR Imaging With Multilevel Distillation. IEEE Signal Processing Letters, 32, 3495-3499. https://doi.org/10.1109/lsp.2025.3601535

Vancouver

Xu J, Cong Y, Deng J, Zeng F, Dai M. Continual NeRF-Based 3D ISAR Imaging With Multilevel Distillation. IEEE Signal Processing Letters. 2025;32:3495-3499. doi:10.1109/lsp.2025.3601535

BibTeX
@article{Xu2025,
  author = {Xu, Jianqiang and Cong, Yulai and Deng, Junyuan and Zeng, Fei and Dai, Mingcheng},
  title = {Continual NeRF-Based 3D ISAR Imaging With Multilevel Distillation},
  journal = {IEEE Signal Processing Letters},
  year = {2025},
  volume = {32},
  pages = {3495--3499},
  doi = {10.1109/lsp.2025.3601535}
}
Kostenlos starten →