Show simple item record

Article

dc.creatorTapiador Morales, Ricardoes
dc.creatorLinares Barranco, Alejandroes
dc.creatorJiménez Fernández, Ángel Franciscoes
dc.creatorJiménez Moreno, Gabrieles
dc.date.accessioned2020-01-13T08:47:10Z
dc.date.available2020-01-13T08:47:10Z
dc.date.issued2018
dc.identifier.citationTapiador Morales, R., Linares Barranco, A., Jiménez Fernández, Á.F. y Jiménez Moreno, G. (2018). Neuromorphic LIF Row-by-Row Multiconvolution Processor for FPGA. IEEE Transactions on Biomedical Circuits and Systems, 13 (1), 159-169.
dc.identifier.issn1932-4545es
dc.identifier.urihttps://hdl.handle.net/11441/91456
dc.description.abstractDeep Learning algorithms have become state-of-theart methods for multiple fields, including computer vision, speech recognition, natural language processing, and audio recognition, among others. In image vision, convolutional neural networks (CNN) stand out. This kind of network is expensive in terms of computational resources due to the large number of operations required to process a frame. In recent years, several frame-based chip solutions to deploy CNN for real time have been developed. Despite the good results in power and accuracy given by these solutions, the number of operations is still high, due the complexity of the current network models. However, it is possible to reduce the number of operations using different computer vision techniques other than frame-based, e.g., neuromorphic event-based techniques. There exist several neuromorphic vision sensors whose pixels detect changes in luminosity. Inspired in the leaky integrate-and-fire (LIF) neuron, we propose in this manuscript an event-based field-programmable gate array (FPGA) multiconvolution system. Its main novelty is the combination of a memory arbiter for efficient memory access to allowrow-by-rowkernel processing. This system is able to convolve 64 filters across multiple kernel sizes, from 1 × 1 to 7 × 7, with latencies of 1.3 μs and 9.01 μs, respectively, generating a continuous flow of output events. The proposed architecture will easily fit spike-based CNNs.es
dc.description.sponsorshipMinisterio de Economía y Competitividad TEC2016-77785-Pes
dc.formatapplication/pdfes
dc.language.isoenges
dc.publisherIEEE Computer Societyes
dc.relation.ispartofIEEE Transactions on Biomedical Circuits and Systems, 13 (1), 159-169.
dc.rightsAttribution-NonCommercial-NoDerivatives 4.0 Internacional*
dc.rights.urihttp://creativecommons.org/licenses/by-nc-nd/4.0/*
dc.subjectAddress-event-representationes
dc.subjectArtificial intelligencees
dc.subjectComputer visiones
dc.subjectConvolutional Neural Networks (CNN)es
dc.subjectDeep learninges
dc.titleNeuromorphic LIF Row-by-Row Multiconvolution Processor for FPGAes
dc.typeinfo:eu-repo/semantics/articlees
dc.type.versioninfo:eu-repo/semantics/submittedVersiones
dc.rights.accessrightsinfo:eu-repo/semantics/openAccesses
dc.contributor.affiliationUniversidad de Sevilla. Departamento de Arquitectura y Tecnología de Computadoreses
dc.relation.projectIDTEC2016-77785-Pes
dc.relation.publisherversionhttps://ieeexplore.ieee.org/document/8526309es
dc.identifier.doi10.1109/TBCAS.2018.2880012es
dc.contributor.groupUniversidad de Sevilla. TEP-108: Robótica y Tecnología de Computadores Aplicada a la Rehabilitaciónes
idus.format.extent11es
dc.journaltitleIEEE Transactions on Biomedical Circuits and Systemses
dc.publication.volumen13es
dc.publication.issue1es
dc.publication.initialPage159es
dc.publication.endPage169es

FilesSizeFormatViewDescription
Neuromorphic LIF Row-by-Row.pdf3.734MbIcon   [PDF] View/Open  

This item appears in the following collection(s)

Show simple item record

Attribution-NonCommercial-NoDerivatives 4.0 Internacional
Except where otherwise noted, this item's license is described as: Attribution-NonCommercial-NoDerivatives 4.0 Internacional