Sign Language Segmentation Using a Transformer-based Approach

Continuous Sign Language Recognition (CSLR), predicting the meaning of the signs in sign language sentences, is one of the current challenges in translation between sign and spoken languages, that would benefit people with hearing impairment. An important limitation of this research field is the lac...

Descripción completa

Detalles Bibliográficos
Autor: Pérez Villegas, Luis Francisco
Tipo de recurso: tesis de maestría
Fecha de publicación:2022
País:España
Institución:Universidad Nacional de Educación a Distancia
Repositorio:e-spacio. Repositorio Institucional de la UNED
Idioma:inglés
OAI Identifier:oai:e-spacio.uned.es:20.500.14468/14662
Acceso en línea:https://hdl.handle.net/20.500.14468/14662
Access Level:acceso abierto
Palabra clave:1203.04 Inteligencia artificial
id ES_eb37e6c27f513582bb9c3fec3be7d8fd
oai_identifier_str oai:e-spacio.uned.es:20.500.14468/14662
network_acronym_str ES
network_name_str España
repository_id_str
spelling Sign Language Segmentation Using a Transformer-based ApproachPérez Villegas, Luis Francisco1203.04 Inteligencia artificialContinuous Sign Language Recognition (CSLR), predicting the meaning of the signs in sign language sentences, is one of the current challenges in translation between sign and spoken languages, that would benefit people with hearing impairment. An important limitation of this research field is the lack of annotated datasets, which could be minimized with Sign Segmentation approaches by automating the costly task of manually annotating the beginning and ending of each sign. The goal of this paper is to study the performance of an architecture which combines I3D CNN extracted features with a transformer-based model called ASFormer which was created specifically for Action Segmentation task. In our approach ASFormer, instead of separating actions in motions is separating signs in a signed speech. Several ablation studies are performed, and it is shown that ASFormer is suitable for segmenting the signs, with a performance near the ones of the state-of-the-art models, confirming the promising benefits of using attention-based approaches in this field.Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia ArtificialSantos, Olga C.e-Spacio UNED20242024-05-2020222022-09-0120222022-09-01master thesishttp://purl.org/coar/resource_type/c_bdccinfo:eu-repo/semantics/masterThesisapplication/pdfhttps://hdl.handle.net/20.500.14468/14662reponame:e-spacio. Repositorio Institucional de la UNEDinstname:Universidad Nacional de Educación a DistanciaInglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccesshttps://creativecommons.org/licenses/by-nc-nd/4.0/deed.esoai:e-spacio.uned.es:20.500.14468/146622026-06-06T12:38:31Z
dc.title.none.fl_str_mv Sign Language Segmentation Using a Transformer-based Approach
title Sign Language Segmentation Using a Transformer-based Approach
spellingShingle Sign Language Segmentation Using a Transformer-based Approach
Pérez Villegas, Luis Francisco
1203.04 Inteligencia artificial
title_short Sign Language Segmentation Using a Transformer-based Approach
title_full Sign Language Segmentation Using a Transformer-based Approach
title_fullStr Sign Language Segmentation Using a Transformer-based Approach
title_full_unstemmed Sign Language Segmentation Using a Transformer-based Approach
title_sort Sign Language Segmentation Using a Transformer-based Approach
dc.creator.none.fl_str_mv Pérez Villegas, Luis Francisco
author Pérez Villegas, Luis Francisco
author_facet Pérez Villegas, Luis Francisco
author_role author
dc.contributor.none.fl_str_mv Santos, Olga C.
e-Spacio UNED
dc.subject.none.fl_str_mv 1203.04 Inteligencia artificial
topic 1203.04 Inteligencia artificial
description Continuous Sign Language Recognition (CSLR), predicting the meaning of the signs in sign language sentences, is one of the current challenges in translation between sign and spoken languages, that would benefit people with hearing impairment. An important limitation of this research field is the lack of annotated datasets, which could be minimized with Sign Segmentation approaches by automating the costly task of manually annotating the beginning and ending of each sign. The goal of this paper is to study the performance of an architecture which combines I3D CNN extracted features with a transformer-based model called ASFormer which was created specifically for Action Segmentation task. In our approach ASFormer, instead of separating actions in motions is separating signs in a signed speech. Several ablation studies are performed, and it is shown that ASFormer is suitable for segmenting the signs, with a performance near the ones of the state-of-the-art models, confirming the promising benefits of using attention-based approaches in this field.
publishDate 2022
dc.date.none.fl_str_mv 2022
2022-09-01
2022
2022-09-01
2024
2024-05-20
dc.type.none.fl_str_mv master thesis
http://purl.org/coar/resource_type/c_bdcc
dc.type.openaire.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
dc.identifier.none.fl_str_mv https://hdl.handle.net/20.500.14468/14662
url https://hdl.handle.net/20.500.14468/14662
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
info:eu-repo/semantics/openAccess
https://creativecommons.org/licenses/by-nc-nd/4.0/deed.es
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
https://creativecommons.org/licenses/by-nc-nd/4.0/deed.es
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia Artificial
publisher.none.fl_str_mv Universidad Nacional de Educación a Distancia (España). Escuela Técnica Superior de Ingeniería Informática. Departamento de Inteligencia Artificial
dc.source.none.fl_str_mv reponame:e-spacio. Repositorio Institucional de la UNED
instname:Universidad Nacional de Educación a Distancia
instname_str Universidad Nacional de Educación a Distancia
reponame_str e-spacio. Repositorio Institucional de la UNED
collection e-spacio. Repositorio Institucional de la UNED
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869423208320466944
score 15.811543

Similares en LA Referencia