AUTHOR=Yang Yajie , Feng Yuxuan , Zhu Li , Fu Haitao , Pan Xin , Jin Chenlei 

TITLE=Feature fusion network based on few-shot fine-grained classification

JOURNAL=Frontiers in Neurorobotics

VOLUME=Volume 17 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/neurorobotics/articles/10.3389/fnbot.2023.1301192

DOI=10.3389/fnbot.2023.1301192

ISSN=1662-5218

ABSTRACT=The objective of few-shot fine-grained learning is to identify subclasses within a primary class using a limited number of labeled samples. However, many current methodologies rely on a single similarity measure, be it global or local. In fine-grained image classification tasks, where the inter-class distance is minimal and the intra-class distance is expansive, relying on a single similarity measure can lead to the omission of either inter-class or intra-class information. We delve into inter-class information through global measures and tap into intra-class information via local measures. In this study, we introduce the Feature Fusion Similarity Network (FFSNet). This model employs global measures to accentuate the differences between classes, while utilizing local measures to consolidate intra-class data. Such an approach enables the model to learn features characterized by pronounced inter-class distances and minimized intra-class distances, even with a limited dataset of fine-grained images. Consequently, this greatly enhances the model's generalization capabilities. Our experimental results indicate that the proposed method stands its ground against other leading algorithms across multiple established fine-grained image benchmark datasets.