Image visual similarity with deep learning: application to a fashion ecommerce company
Autor(a) principal: | |
---|---|
Data de Publicação: | 2017 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
Texto Completo: | https://repositorio-aberto.up.pt/handle/10216/106962 |
Resumo: | Deep learning is a very trendy topic now, showing high accuracy in image based systems that can go from image segmentation to object detection and image retrieval. Because of this, multiple researchers and companies have been building and sharing work in the community, including pre-trained convolutional neural networks, available for public use. This work follows the trend and delivers an experimental study using deep learning for building a visually similar image retrieval application, comparing three different convolutional neural architectures for feature extraction and six distance indexes for similarity calculation in a real-world image retrieval problem, using real data from a fashion e-commerce platform from Morocco. After testing all the different combinations, we can conclude that for this dataset, Vgg19 combined with a correlation coefficient for similarity calculation is the tuple that best maximizes the similarity between a search image and its retrieved neighbors. |
id |
RCAP_6f440acf4f16f47b45a326dc7a71ecbf |
---|---|
oai_identifier_str |
oai:repositorio-aberto.up.pt:10216/106962 |
network_acronym_str |
RCAP |
network_name_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository_id_str |
7160 |
spelling |
Image visual similarity with deep learning: application to a fashion ecommerce companyEconomia e gestãoEconomics and BusinessDeep learning is a very trendy topic now, showing high accuracy in image based systems that can go from image segmentation to object detection and image retrieval. Because of this, multiple researchers and companies have been building and sharing work in the community, including pre-trained convolutional neural networks, available for public use. This work follows the trend and delivers an experimental study using deep learning for building a visually similar image retrieval application, comparing three different convolutional neural architectures for feature extraction and six distance indexes for similarity calculation in a real-world image retrieval problem, using real data from a fashion e-commerce platform from Morocco. After testing all the different combinations, we can conclude that for this dataset, Vgg19 combined with a correlation coefficient for similarity calculation is the tuple that best maximizes the similarity between a search image and its retrieved neighbors.2017-07-252017-07-25T00:00:00Zinfo:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesisapplication/pdfhttps://repositorio-aberto.up.pt/handle/10216/106962TID:201929163engRui Pedro da Silva Rodrigues Machadoinfo:eu-repo/semantics/openAccessreponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãoinstacron:RCAAP2023-11-29T14:45:32Zoai:repositorio-aberto.up.pt:10216/106962Portal AgregadorONGhttps://www.rcaap.pt/oai/openaireopendoar:71602024-03-20T00:07:55.717937Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informaçãofalse |
dc.title.none.fl_str_mv |
Image visual similarity with deep learning: application to a fashion ecommerce company |
title |
Image visual similarity with deep learning: application to a fashion ecommerce company |
spellingShingle |
Image visual similarity with deep learning: application to a fashion ecommerce company Rui Pedro da Silva Rodrigues Machado Economia e gestão Economics and Business |
title_short |
Image visual similarity with deep learning: application to a fashion ecommerce company |
title_full |
Image visual similarity with deep learning: application to a fashion ecommerce company |
title_fullStr |
Image visual similarity with deep learning: application to a fashion ecommerce company |
title_full_unstemmed |
Image visual similarity with deep learning: application to a fashion ecommerce company |
title_sort |
Image visual similarity with deep learning: application to a fashion ecommerce company |
author |
Rui Pedro da Silva Rodrigues Machado |
author_facet |
Rui Pedro da Silva Rodrigues Machado |
author_role |
author |
dc.contributor.author.fl_str_mv |
Rui Pedro da Silva Rodrigues Machado |
dc.subject.por.fl_str_mv |
Economia e gestão Economics and Business |
topic |
Economia e gestão Economics and Business |
description |
Deep learning is a very trendy topic now, showing high accuracy in image based systems that can go from image segmentation to object detection and image retrieval. Because of this, multiple researchers and companies have been building and sharing work in the community, including pre-trained convolutional neural networks, available for public use. This work follows the trend and delivers an experimental study using deep learning for building a visually similar image retrieval application, comparing three different convolutional neural architectures for feature extraction and six distance indexes for similarity calculation in a real-world image retrieval problem, using real data from a fashion e-commerce platform from Morocco. After testing all the different combinations, we can conclude that for this dataset, Vgg19 combined with a correlation coefficient for similarity calculation is the tuple that best maximizes the similarity between a search image and its retrieved neighbors. |
publishDate |
2017 |
dc.date.none.fl_str_mv |
2017-07-25 2017-07-25T00:00:00Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://repositorio-aberto.up.pt/handle/10216/106962 TID:201929163 |
url |
https://repositorio-aberto.up.pt/handle/10216/106962 |
identifier_str_mv |
TID:201929163 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.source.none.fl_str_mv |
reponame:Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) instname:Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação instacron:RCAAP |
instname_str |
Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
instacron_str |
RCAAP |
institution |
RCAAP |
reponame_str |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
collection |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) |
repository.name.fl_str_mv |
Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos) - Agência para a Sociedade do Conhecimento (UMIC) - FCT - Sociedade da Informação |
repository.mail.fl_str_mv |
|
_version_ |
1799136005128716289 |