EUVIMEDEUVIMED entdecken
Crossref

Self-Supervised Image Segmentation Using Meta-Learning and Multi-Backbone Feature Fusion

Muhammad Shahroz Ajmal iD, Guohua Geng iD, Xiaofeng Wang iD, Mohsin Ashraf iD

International Journal of Neural Systems · 2025

DOI: 10.1142/s0129065725500121 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Neural Systems, 2025. Verfasst von Muhammad Shahroz Ajmal und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Information Science and Technology, Northwest University, Xi'an, 710069, P. R. China. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted, Supervised Machine Learning. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 38 Literaturangaben. Gefördert von: General Projects of Shaanxi Provincial Key R&D Plan - Industrial Field, Xi’an Science and Technology Plan Project, National Natural Science Foundation of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Few-shot segmentation (FSS) aims to reduce the need for manual annotation, which is both expensive and time-consuming. While FSS enhances model generalization to new concepts with only limited test samples, it still relies on a substantial amount of labeled training data for base classes. To address these issues, we propose a multi-backbone few shot segmentation (MBFSS) method. This self-supervised FSS technique utilizes unsupervised saliency for pseudo-labeling, allowing the model to be trained on unlabeled data. In addition, it integrates features from multiple backbones (ResNet, ResNeXt, and PVT v2) to generate a richer feature representation than a single backbone. Through extensive experimentation on PASCAL-5i and COCO-20i, our method achieves 54.3% and 25.1% on one-shot segmentation, exceeding the baseline methods by 13.5% and 4%, respectively. These improvements significantly enhance the model’s performance in real-world applications with negligible labeling effort.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Image Processing, Computer-AssistedSupervised Machine LearningHumansAlgorithmsNeural Networks, Computer

Stichwörter der Autor:innen

Few-shotfeature fusionmulti-backboneself-supervisionsemantic segmentation

Autor:innen und Institutionen

  • School of Information Science and Technology, Northwest University, Xi'an, 710069, P. R. China.
  • Department of Computer Sciences, Faculty of Information Technology and Computer Sciences, University of Central Punjab, Lahore 54590, Pakistan.

ORCID-Profile: Muhammad Shahroz Ajmal, Guohua Geng, Xiaofeng Wang, Mohsin Ashraf

Förderung

  • General Projects of Shaanxi Provincial Key R&D Plan - Industrial Field
  • Xi’an Science and Technology Plan Project
  • National Natural Science Foundation of China

Zitieren

APA

Ajmal, M. S., Geng, G., Wang, X., & Ashraf, M. (2025). Self-Supervised Image Segmentation Using Meta-Learning and Multi-Backbone Feature Fusion. International Journal of Neural Systems. https://doi.org/10.1142/s0129065725500121

Vancouver

Ajmal MS, Geng G, Wang X, Ashraf M. Self-Supervised Image Segmentation Using Meta-Learning and Multi-Backbone Feature Fusion. International Journal of Neural Systems. 2025. doi:10.1142/s0129065725500121

BibTeX
@article{Ajmal2025,
  author = {Ajmal, Muhammad Shahroz and Geng, Guohua and Wang, Xiaofeng and Ashraf, Mohsin},
  title = {Self-Supervised Image Segmentation Using Meta-Learning and Multi-Backbone Feature Fusion},
  journal = {International Journal of Neural Systems},
  year = {2025},
  doi = {10.1142/s0129065725500121}
}
Kostenlos starten →