EUVIMEDEUVIMED entdecken
Crossref

GAIN: Global-Atomic INteraction Graph for Few-Shot Class-Incremental Learning

Fan Lyu, Linglan Zhao, Chengyan Liu, Yinying Mei, Zhang Zhang, Baoqing Yu, Tao Zhang, Fuyuan Hu, Liang Wang

IEEE Transactions on Circuits and Systems for Video Technology · 2026-07 · 36 (7) · 10019-10035

DOI: 10.1109/tcsvt.2026.3673928 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-07. Verfasst von Fan Lyu und 8 weiteren Autor:innen. Die Arbeit nennt 72 Literaturangaben. Gefördert von: National Science and Technology Major Project, National Natural Science Foundation of China, Postdoctoral Fellowship Program of China Postdoctoral Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Few-shot class-incremental learning (FSCIL) aims to develop models that can continually learn novel classes from only a few training samples while retaining knowledge of previously learned classes. The primary challenges of FSCIL lie in the limited available data in novel sessions, which can cause severe overfitting and exacerbate the catastrophic forgetting of novel classes. Prior studies have demonstrated the effectiveness of modeling graph relationships in few-shot learning, which can alleviate overfitting through feature aggregation. Building on these insights, we extend the concept to the incremental learning scenario and propose a Global-Atomic INteraction (GAIN) graph learning framework for FSCIL. Specifically, we introduce an Atomic-session Graph Network (AGN) to capture relationships among samples within each session, enabling the aggregation of similar samples and the extraction of more representative global-session features. Additionally, we design a Global-session Graph Network (GGN) that connects global-session features across base and novel classes, facilitating knowledge transfer between sessions and enhancing overall learning performance. Finally, we utilize a multi-stage strategy for training the GAIN model, which suppresses the training challenges posed by limited data in incremental learning sessions. The multi-stage training strategy is designed to build a GAIN graph from base to novel sessions and improve the capacity via an extra pseudo-incremental stage. Extensive experiments on three widely used FSCIL benchmarks demonstrate that our approach consistently outperforms existing methods and establishes new state-of-the-art performance. Our code will be released upon acceptance.

Förderung

  • National Science and Technology Major Project
  • National Natural Science Foundation of China
  • Postdoctoral Fellowship Program of China Postdoctoral Science Foundation
  • CPSF

Zitieren

APA

Lyu, F., Zhao, L., Liu, C., Mei, Y., Zhang, Z., Yu, B., Zhang, T., Hu, F., & Wang, L. (2026). GAIN: Global-Atomic INteraction Graph for Few-Shot Class-Incremental Learning. IEEE Transactions on Circuits and Systems for Video Technology, 36(7), 10019-10035. https://doi.org/10.1109/tcsvt.2026.3673928

Vancouver

Lyu F, Zhao L, Liu C, Mei Y, Zhang Z, Yu B, et al. GAIN: Global-Atomic INteraction Graph for Few-Shot Class-Incremental Learning. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(7):10019-10035. doi:10.1109/tcsvt.2026.3673928

BibTeX
@article{Lyu2026,
  author = {Lyu, Fan and Zhao, Linglan and Liu, Chengyan and Mei, Yinying and Zhang, Zhang and Yu, Baoqing and Zhang, Tao and Hu, Fuyuan and Wang, Liang},
  title = {GAIN: Global-Atomic INteraction Graph for Few-Shot Class-Incremental Learning},
  journal = {IEEE Transactions on Circuits and Systems for Video Technology},
  year = {2026},
  volume = {36},
  number = {7},
  pages = {10019--10035},
  doi = {10.1109/tcsvt.2026.3673928}
}
Kostenlos starten →