EUVIMEDEUVIMED entdecken
Crossref

Adaptive Mixture-of-Experts Distillation for Cross-Satellite Generalizable Incremental Remote Sensing Scene Classification

Yimin Fu, Runqing Yang, Zhunga Liu, Michael K. Ng

IEEE Transactions on Circuits and Systems for Video Technology · 2026-01 · 36 (1) · 233-247

DOI: 10.1109/tcsvt.2025.3598274 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-01. Verfasst von Yimin Fu und 3 weiteren Autor:innen. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 96 Literaturangaben. Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China, RGC GRF.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Incremental learning aims to continuously acquire new knowledge from data streams while maintaining previously learned knowledge. Existing incremental learning methods typically assume that the training (source domain) and testing (target domain) data are identically distributed. However, differences in sensor parameters and imaging conditions inevitably lead to distribution gaps between data collected from different satellites (domains). The ensuing domain shift problem substantially impairs the generalization of continuously learned knowledge from source domains to unseen ones. To tackle this problem, we propose adaptive mixture-of-experts distillation (AMoED) for cross-satellite generalizable incremental remote sensing scene classification (CS-GIRSSC). Specifically, AMoED adopts a high-level semantic learning pipeline, in which new knowledge is acquired through the coordinated guidance of multiple domain-specific experts, rather than directly from raw data. This pipeline prevents the model from being exposed to large volumes of newly emerging data, thereby alleviating the erasure of previous knowledge when adapting to new data distributions. Besides, the adaptive mixture of domain-specific experts facilitates the formation of universal class concepts, which exhibit strong generalizability across different domains. During the learning process, an equi-partite subset is constructed for knowledge acquisition and consolidation, accompanied by a shallow style-mixing operation to mitigate the interference of domain discrepancies. Extensive experiments are conducted on four remote sensing scene classification datasets, and the proposed method consistently achieves state-of-the-art performance across various scenarios and settings. The code is released at https://github.com/fuyimin96/AMoED

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China
  • RGC GRF
  • Joint NSFC and RGC

Zitieren

APA

Fu, Y., Yang, R., Liu, Z., & Ng, M. K. (2026). Adaptive Mixture-of-Experts Distillation for Cross-Satellite Generalizable Incremental Remote Sensing Scene Classification. IEEE Transactions on Circuits and Systems for Video Technology, 36(1), 233-247. https://doi.org/10.1109/tcsvt.2025.3598274

Vancouver

Fu Y, Yang R, Liu Z, Ng MK. Adaptive Mixture-of-Experts Distillation for Cross-Satellite Generalizable Incremental Remote Sensing Scene Classification. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(1):233-247. doi:10.1109/tcsvt.2025.3598274

BibTeX
@article{Fu2026,
  author = {Fu, Yimin and Yang, Runqing and Liu, Zhunga and Ng, Michael K.},
  title = {Adaptive Mixture-of-Experts Distillation for Cross-Satellite Generalizable Incremental Remote Sensing Scene Classification},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2026},
  volume = {36},
  number = {1},
  pages = {233--247},
  doi = {10.1109/tcsvt.2025.3598274}
}
Kostenlos starten →