Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet
- Autores
- Pascal, Andrés; Planas, Adrián; Vidal, Zoe Florencia; Bonti, Agustina; Tonelotto, Lucas; Castiglioni, León
- Año de publicación
- 2024
- Idioma
- inglés
- Tipo de recurso
- documento de conferencia
- Estado
- versión publicada
- Descripción
- This study evaluates the use of feature extraction with a pre-trained ResNet50 model for similarity search tasks. We employed transfer learning from both initial and intermediate layers of ResNet50 and applied a robust preprocessing approach, including resizing and Gaussian blur, to optimize feature extraction. The method achieved over 90% accuracy in nearest neighbor (NN) and k-Nearest Neighbors (k=3 and k=5) searches for both logos and paintings datasets. Notably, our approach does not require fine-tuning, which is advantageous when only one instance of each element is available. Overall, the method is effective and practical for similarity search applications.
Red de Universidades con Carreras en Informática - Materia
-
Ciencias Informáticas
Image Similarity Search
Feature Extraction
CNNs
Transfer Learning
Logos
Paintings - Nivel de accesibilidad
- acceso abierto
- Condiciones de uso
- http://creativecommons.org/licenses/by-nc-sa/4.0/
- Repositorio
.jpg)
- Institución
- Universidad Nacional de La Plata
- OAI Identificador
- oai:sedici.unlp.edu.ar:10915/176283
Ver los metadatos del registro completo
| id |
SEDICI_c7200749de5cb328b8a3e0411f9a8de7 |
|---|---|
| oai_identifier_str |
oai:sedici.unlp.edu.ar:10915/176283 |
| network_acronym_str |
SEDICI |
| repository_id_str |
1329 |
| network_name_str |
SEDICI (UNLP) |
| spelling |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNetPascal, AndrésPlanas, AdriánVidal, Zoe FlorenciaBonti, AgustinaTonelotto, LucasCastiglioni, LeónCiencias InformáticasImage Similarity SearchFeature ExtractionCNNsTransfer LearningLogosPaintingsThis study evaluates the use of feature extraction with a pre-trained ResNet50 model for similarity search tasks. We employed transfer learning from both initial and intermediate layers of ResNet50 and applied a robust preprocessing approach, including resizing and Gaussian blur, to optimize feature extraction. The method achieved over 90% accuracy in nearest neighbor (NN) and k-Nearest Neighbors (k=3 and k=5) searches for both logos and paintings datasets. Notably, our approach does not require fine-tuning, which is advantageous when only one instance of each element is available. Overall, the method is effective and practical for similarity search applications.Red de Universidades con Carreras en Informática2024-10info:eu-repo/semantics/conferenceObjectinfo:eu-repo/semantics/publishedVersionObjeto de conferenciahttp://purl.org/coar/resource_type/c_5794info:ar-repo/semantics/documentoDeConferenciaapplication/pdf82-93http://sedici.unlp.edu.ar/handle/10915/176283enginfo:eu-repo/semantics/altIdentifier/isbn/978-950-34-2428-5info:eu-repo/semantics/reference/hdl/10915/172755info:eu-repo/semantics/openAccesshttp://creativecommons.org/licenses/by-nc-sa/4.0/Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)reponame:SEDICI (UNLP)instname:Universidad Nacional de La Platainstacron:UNLP2025-09-29T11:47:28Zoai:sedici.unlp.edu.ar:10915/176283Institucionalhttp://sedici.unlp.edu.ar/Universidad públicaNo correspondehttp://sedici.unlp.edu.ar/oai/snrdalira@sedici.unlp.edu.arArgentinaNo correspondeNo correspondeNo correspondeopendoar:13292025-09-29 11:47:28.755SEDICI (UNLP) - Universidad Nacional de La Platafalse |
| dc.title.none.fl_str_mv |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| title |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| spellingShingle |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet Pascal, Andrés Ciencias Informáticas Image Similarity Search Feature Extraction CNNs Transfer Learning Logos Paintings |
| title_short |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| title_full |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| title_fullStr |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| title_full_unstemmed |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| title_sort |
Image Feature Extraction for Similarity Searching Using Transfer Learning with ResNet |
| dc.creator.none.fl_str_mv |
Pascal, Andrés Planas, Adrián Vidal, Zoe Florencia Bonti, Agustina Tonelotto, Lucas Castiglioni, León |
| author |
Pascal, Andrés |
| author_facet |
Pascal, Andrés Planas, Adrián Vidal, Zoe Florencia Bonti, Agustina Tonelotto, Lucas Castiglioni, León |
| author_role |
author |
| author2 |
Planas, Adrián Vidal, Zoe Florencia Bonti, Agustina Tonelotto, Lucas Castiglioni, León |
| author2_role |
author author author author author |
| dc.subject.none.fl_str_mv |
Ciencias Informáticas Image Similarity Search Feature Extraction CNNs Transfer Learning Logos Paintings |
| topic |
Ciencias Informáticas Image Similarity Search Feature Extraction CNNs Transfer Learning Logos Paintings |
| dc.description.none.fl_txt_mv |
This study evaluates the use of feature extraction with a pre-trained ResNet50 model for similarity search tasks. We employed transfer learning from both initial and intermediate layers of ResNet50 and applied a robust preprocessing approach, including resizing and Gaussian blur, to optimize feature extraction. The method achieved over 90% accuracy in nearest neighbor (NN) and k-Nearest Neighbors (k=3 and k=5) searches for both logos and paintings datasets. Notably, our approach does not require fine-tuning, which is advantageous when only one instance of each element is available. Overall, the method is effective and practical for similarity search applications. Red de Universidades con Carreras en Informática |
| description |
This study evaluates the use of feature extraction with a pre-trained ResNet50 model for similarity search tasks. We employed transfer learning from both initial and intermediate layers of ResNet50 and applied a robust preprocessing approach, including resizing and Gaussian blur, to optimize feature extraction. The method achieved over 90% accuracy in nearest neighbor (NN) and k-Nearest Neighbors (k=3 and k=5) searches for both logos and paintings datasets. Notably, our approach does not require fine-tuning, which is advantageous when only one instance of each element is available. Overall, the method is effective and practical for similarity search applications. |
| publishDate |
2024 |
| dc.date.none.fl_str_mv |
2024-10 |
| dc.type.none.fl_str_mv |
info:eu-repo/semantics/conferenceObject info:eu-repo/semantics/publishedVersion Objeto de conferencia http://purl.org/coar/resource_type/c_5794 info:ar-repo/semantics/documentoDeConferencia |
| format |
conferenceObject |
| status_str |
publishedVersion |
| dc.identifier.none.fl_str_mv |
http://sedici.unlp.edu.ar/handle/10915/176283 |
| url |
http://sedici.unlp.edu.ar/handle/10915/176283 |
| dc.language.none.fl_str_mv |
eng |
| language |
eng |
| dc.relation.none.fl_str_mv |
info:eu-repo/semantics/altIdentifier/isbn/978-950-34-2428-5 info:eu-repo/semantics/reference/hdl/10915/172755 |
| dc.rights.none.fl_str_mv |
info:eu-repo/semantics/openAccess http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) |
| eu_rights_str_mv |
openAccess |
| rights_invalid_str_mv |
http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) |
| dc.format.none.fl_str_mv |
application/pdf 82-93 |
| dc.source.none.fl_str_mv |
reponame:SEDICI (UNLP) instname:Universidad Nacional de La Plata instacron:UNLP |
| reponame_str |
SEDICI (UNLP) |
| collection |
SEDICI (UNLP) |
| instname_str |
Universidad Nacional de La Plata |
| instacron_str |
UNLP |
| institution |
UNLP |
| repository.name.fl_str_mv |
SEDICI (UNLP) - Universidad Nacional de La Plata |
| repository.mail.fl_str_mv |
alira@sedici.unlp.edu.ar |
| _version_ |
1844616338260099072 |
| score |
13.070432 |