Saved in:
Bibliographic Details
Main Authors: Artiaga, Keren, Lynch, Conor, Afli, Haithem, Hasanuzzaman, Mohammed
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.03316
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911483460321280
author Artiaga, Keren
Lynch, Conor
Afli, Haithem
Hasanuzzaman, Mohammed
author_facet Artiaga, Keren
Lynch, Conor
Afli, Haithem
Hasanuzzaman, Mohammed
contents Most sign language recognition research relies on Transfer Learning (TL) from vision-based datasets such as ImageNet. Some extend this to alternatively available language datasets, often focusing on signs with cross-linguistic similarities. This body of work examines the necessity of these likenesses on effective knowledge transfer by comparing TL performance between iconic signs of two different sign language pairs: Chinese to Arabic and Greek to Flemish. Google Mediapipe was utilised as an input feature extractor, enabling spatial information of these signs to be processed with a Multilayer Perceptron architecture and the temporal information with a Gated Recurrent Unit. Experimental results showed a 7.02% improvement for Arabic and 1.07% for Flemish when conducting iconic TL from Chinese and Greek respectively.
format Preprint
id arxiv_https___arxiv_org_abs_2603_03316
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle The Influence of Iconicity in Transfer Learning for Sign Language Recognition
Artiaga, Keren
Lynch, Conor
Afli, Haithem
Hasanuzzaman, Mohammed
Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
I.2.10; I.5.4
Most sign language recognition research relies on Transfer Learning (TL) from vision-based datasets such as ImageNet. Some extend this to alternatively available language datasets, often focusing on signs with cross-linguistic similarities. This body of work examines the necessity of these likenesses on effective knowledge transfer by comparing TL performance between iconic signs of two different sign language pairs: Chinese to Arabic and Greek to Flemish. Google Mediapipe was utilised as an input feature extractor, enabling spatial information of these signs to be processed with a Multilayer Perceptron architecture and the temporal information with a Gated Recurrent Unit. Experimental results showed a 7.02% improvement for Arabic and 1.07% for Flemish when conducting iconic TL from Chinese and Greek respectively.
title The Influence of Iconicity in Transfer Learning for Sign Language Recognition
topic Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
I.2.10; I.5.4
url https://arxiv.org/abs/2603.03316