Enhanced Few-Shot Learning for Indonesian Sign Language with Prototypical Networks Approach

Irma Permata Sari; Fuad Mumtas; Z.E. Ferdi Fauzan Putra; Ressy Dwitias Sari; Ati Zaidiah; Mayanda Mega Snatoni

doi:10.1109/ICIMCIS60089.2023.10349031

Enhanced Few-Shot Learning for Indonesian Sign Language with Prototypical Networks Approach

Closed

Irma Permata Sari, Fuad Mumtas, Z.E. Ferdi Fauzan Putra, Ressy Dwitias Sari, Ati Zaidiah, Mayanda Mega Snatoni

2023 2023 International Conference on Informatics, Multimedia, Cyber and Information Systems, ICIMCIS 2023 Conference paper Cited by 7

Abstract

The issue of few-shot learning (FSL), which requires using a small amount of training data to categorize instances from classes that have never been encountered before. We concentrate on categorizing images in Indonesian Sign Language using a few-shot image classification method utilizing Prototypical Networks. To accomplish this, we used a neural network to apply a non-linear transformation to the input data, with each class being represented by the means of its support set within the embedding space. The theory of few-shot learning serves as the foundation for the paradigm used in this investigation. According to our experimental results, the Shufflenet-V2 convolutional network model for Prototypical Networks produced the best accuracy, reaching 96.75%. © 2023 IEEE.

Affiliations

Universitas Negeri Jakarta, Information System and Technology Department, Jakarta, Indonesia; Universitas Negeri Jakarta, Informatic Education Department, Jakarta, Indonesia; Universitas Pembangunan Nasional Veteran Jakarta, Information System Departement, Jakarta, Indonesia; Universitas Pembangunan Nasional Veteran Jakarta, Informatic Departement, Jakarta, Indonesia

View full text (DOI) →

Research at a Glance

Research at a Glance

Topics & SDG Alignment

Collaboration

Author Profile (Selected)

References Overview

Journal & Source

Metadata & Integrity