Sketch-QNet: A quadruplet convnet for color sketch-based image retrieval

Anibal Fuentes, Jose M. Saavedra

Producción científica: Capítulo del libro/informe/acta de congresoContribución a la conferenciarevisión exhaustiva

10 Citas (Scopus)

Resumen

Architectures based on siamese networks with triplet loss have shown outstanding performance on the image-based similarity search problem. This approach attempts to discriminate between positive (relevant) and negative (irrelevant) items. However, it undergoes a critical weakness. Given a query, it cannot discriminate weakly relevant items, for instance, items of the same type but different color or texture as the given query, which could be a serious limitation for many real-world search applications. Therefore, in this work, we present a quadruplet-based architecture that overcomes the aforementioned weakness. Moreover, we present an instance of this quadruplet network, which we call Sketch-QNet, to deal with the color sketch-based image retrieval (CSBIR) problem, achieving new state-of-the-art results.

Idioma originalInglés
Título de la publicación alojadaProceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
EditorialIEEE Computer Society
Páginas2134-2141
Número de páginas8
ISBN (versión digital)9781665448994
DOI
EstadoPublicada - jun. 2021
Publicado de forma externa
Evento2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021 - Virtual, Online, Estados Unidos
Duración: 19 jun. 202125 jun. 2021

Serie de la publicación

NombreIEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
ISSN (versión impresa)2160-7508
ISSN (versión digital)2160-7516

Conferencia

Conferencia2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
País/TerritorioEstados Unidos
CiudadVirtual, Online
Período19/06/2125/06/21

Nota bibliográfica

Publisher Copyright:
© 2021 IEEE.

Huella

Profundice en los temas de investigación de 'Sketch-QNet: A quadruplet convnet for color sketch-based image retrieval'. En conjunto forman una huella única.

Citar esto