Class-Irrelevant Feature Removal for Few-Shot Image Classification

Most existing few-shot image classification methods employ global pooling to aggregate class-relevant local features in a data-drive manner. Due to the difficulty and inaccuracy in locating class-relevant regions in complex scenarios, as well as the large semantic diversity of local features, the cl...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	IEEE transaction on neural networks and learning systems 2024-07, Vol.PP, p.1-15
Hauptverfasser:	Hao, Fusheng, Liu, Liu, Wu, Fuxiang, Zhang, Qieshi, Cheng, Jun
Format:	Artikel
Sprache:	eng
Schlagworte:	Class-irrelevant feature removal (CIFR) Computational modeling deep models few-shot image classification Image classification Image representation Loss measurement Measurement metric learning Semantics Task analysis
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!