English   español  
Por favor, use este identificador para citar o enlazar a este item: http://hdl.handle.net/10261/171796
COMPARTIR / IMPACTO:
Estadísticas
logo share SHARE   Add this article to your Mendeley library MendeleyBASE
Visualizar otros formatos: MARC | Dublin Core | RDF | ORE | MODS | METS | DIDL
Exportar a otros formatos:
Título

Optimum Network/Framework Selection from High-Level Specifications in Embedd

AutorVelasco-Montero, Delia; Fernández-Berni, J. ; Carmona-Galán, R. ; Rodríguez-Vázquez, Ángel
Palabras claveDeep Learning
Convolutional Neural Networks
Embedded Vision
Performance
High-Level Specifications
Fecha de publicaciónoct-2018
EditorSpringer
CitaciónLecture Notes in Computational Science and Engineering, 11182: 369-379 (2018)
ResumenThis paper benchmarks 16 combinations of popular Deep Neural Networks for 1000-category image recognition and Deep Learn- ing frameworks on an embedded platform. A Figure of Merit based on high-level specifications is introduced. By sweeping the relative weight of accuracy, throughput and power consumption on global performance, we demonstrate that only a reduced set of the analyzed combinations must actually be considered for real deployment. We also report the op- timum network/framework selection for all possible application scenarios de ned in those terms, i.e. weighted balance of the aforementioned pa- rameters. Our approach can be extended to other networks, frameworks and performance parameters, thus supporting system-level design deci- sions in the ever-changing ecosystem of Deep Learning technology
Descripciónin Advanced Concepts for Intelligent Vision Systems (ACIVS), Poitiers, France, September 2018, ISBN 978-3-030-01448-3,
URIhttp://hdl.handle.net/10261/171796
ISSN978-3-030-01448-3
Aparece en las colecciones: (IMSE-CNM) Artículos
Ficheros en este ítem:
Fichero Descripción Tamaño Formato  
ACIVS_2018.pdf Embargado hasta 1 de octubre de 2019420,95 kBAdobe PDFVista previa
Visualizar/Abrir     Petición de una copia
Mostrar el registro completo
 


NOTA: Los ítems de Digital.CSIC están protegidos por copyright, con todos los derechos reservados, a menos que se indique lo contrario.