EUVIMEDEUVIMED entdecken
Crossref

Optimising few-shot class-incremental learning for fine-grained visual recognition

Yimin Yin, Haoling Liu, Sihang Xu, Renye Zhang, Jinghua Zhang

International Journal of Multimedia Information Retrieval · 2026-06 · 15 (2)

DOI: 10.1007/s13735-026-00409-8 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Multimedia Information Retrieval, online erschienen am 04.06.2026. Verfasst von Yimin Yin und 4 weiteren Autor:innen. Die Arbeit nennt 52 Literaturangaben. Gefördert von: The Excellent Youth Project of Hunan Provincial Education Department.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional Deep Learning models often rely on extensive training datasets and struggle in dynamic environments. Few-Shot Class-Incremental Learning (FSCIL) addresses these challenges by enabling models to learn new classes with limited samples while retaining old knowledge. However, FSCIL models typically perform well on coarse-grained datasets but falter on fine-grained ones due to high inter-class similarity and limited samples for new classes. This paper introduces a novel framework that mitigates catastrophic forgetting by leveraging the generalization capabilities of the CLIP image encoder through Cosine-based Distillation. Additionally, we employ supervised contrastive learning and a forward-compatible strategy to address the challenges specific to Fine-Grained FSCIL (FG-FSCIL). Our method demonstrates state-of-the-art performance on four fine-grained datasets (CUB200, Stanford Dogs, Stanford Cars, and Aircraft), achieving significant improvements over existing methods. The code and datasets are available at https://github.com/yahuuuuuuu/Fine-Grained-FSCIL.

Förderung

  • The Excellent Youth Project of Hunan Provincial Education Department

Zitieren

APA

Yin, Y., Liu, H., Xu, S., Zhang, R., & Zhang, J. (2026). Optimising few-shot class-incremental learning for fine-grained visual recognition. International Journal of Multimedia Information Retrieval, 15(2). https://doi.org/10.1007/s13735-026-00409-8

Vancouver

Yin Y, Liu H, Xu S, Zhang R, Zhang J. Optimising few-shot class-incremental learning for fine-grained visual recognition. International Journal of Multimedia Information Retrieval. 2026;15(2). doi:10.1007/s13735-026-00409-8

BibTeX
@article{Yin2026,
  author = {Yin, Yimin and Liu, Haoling and Xu, Sihang and Zhang, Renye and Zhang, Jinghua},
  title = {Optimising few-shot class-incremental learning for fine-grained visual recognition},
  journal = {International Journal of Multimedia Information Retrieval},
  year = {2026},
  volume = {15},
  number = {2},
  doi = {10.1007/s13735-026-00409-8}
}
Kostenlos starten →