MergedNET: A simple approach for one-shot learning in siamese networks based on similarity layers

Atanbori, John and Rose, Samuel (2022) MergedNET: A simple approach for one-shot learning in siamese networks based on similarity layers. Neurocomputing, 509 . pp. 1-10. ISSN 0925-2312

Full content URL: https://doi.org/10.1016/j.neucom.2022.08.070

Documents
MergedNET: A simple approach for one-shot learning in siamese networks based on similarity layers
Open Access manuscript
[img]
[Download]
[img]
Preview
PDF
1-s2.0-S0925231222010529-main.pdf - Whole Document
Available under License Creative Commons Attribution 4.0 International.

2MB
Item Type:Article
Item Status:Live Archive

Abstract

Classifiers trained on disjointed classes with few labelled data points are used in one-shot learning to identify visual concepts from other classes. Recently, Siamese networks and similarity layers have been used to solve the one-shot learning problem, achieving state-of-the-art performance on visual-character recognition datasets. Various techniques have been developed over the years to improve the performance of these networks on fine-grained image classification datasets. They focused primarily on improving the loss and activation functions, augmenting visual features, employing multiscale metric learning, and pre-training and fine-tuning the backbone network. We investigate similarity layers for one-shot learning tasks and propose two frameworks for combining these layers into a MergedNet network. On all four datasets used in our experiment, MergedNet outperformed the baselines based on classification accuracy, and it generalises to other datasets when trained on miniImageNet.

Keywords:Convolutional Neural Networks (CNN), Similarity layers, Siamese networks, One-shot learning
Subjects:G Mathematical and Computer Sciences > G400 Computer Science
G Mathematical and Computer Sciences > G760 Machine Learning
G Mathematical and Computer Sciences > G740 Computer Vision
Divisions:College of Science > School of Computer Science
Related URLs:
ID Code:52732
Deposited On:19 Dec 2022 09:36

Repository Staff Only: item control page