TUTCRIS - Tampereen teknillinen yliopisto

TUTCRIS

Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification

Tutkimustuotosvertaisarvioitu

Standard

Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification. / Xie, Renjie; Huttunen, Heikki; Lin, Shuoxin; Bhattacharyya, Shuvra; Takala, Jarmo.

2016 24th European Signal Processing Conference (EUSIPCO). IEEE, 2016. s. 1862-1866.

Tutkimustuotosvertaisarvioitu

Harvard

Xie, R, Huttunen, H, Lin, S, Bhattacharyya, S & Takala, J 2016, Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification. julkaisussa 2016 24th European Signal Processing Conference (EUSIPCO). IEEE, Sivut 1862-1866, EUROPEAN SIGNAL PROCESSING CONFERENCE, 1/01/00.

APA

Xie, R., Huttunen, H., Lin, S., Bhattacharyya, S., & Takala, J. (2016). Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification. teoksessa 2016 24th European Signal Processing Conference (EUSIPCO) (Sivut 1862-1866). IEEE.

Vancouver

Xie R, Huttunen H, Lin S, Bhattacharyya S, Takala J. Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification. julkaisussa 2016 24th European Signal Processing Conference (EUSIPCO). IEEE. 2016. s. 1862-1866

Author

Xie, Renjie ; Huttunen, Heikki ; Lin, Shuoxin ; Bhattacharyya, Shuvra ; Takala, Jarmo. / Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification. 2016 24th European Signal Processing Conference (EUSIPCO). IEEE, 2016. Sivut 1862-1866

Bibtex - Lataa

@inproceedings{65c052e9dc49467b95f0edf5def05ded,
title = "Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification",
abstract = "Deep learning has attracted great research interest in recent years in many signal processing application areas. However, investigation of deep learning implementations in highly resource-constrained contexts has been relatively unexplored due to the large computational requirements involved. In this paper, we investigate the implementation of a deep learning application for vehicle classification on multicore platforms with limited numbers of available processor cores. We apply model-based design methods based on signal processing oriented dataflow models of computation, and using the resulting dataflow representations, we apply various design optimizations to derive efficient implementations on three different multicore platforms. Using model-based design techniques throughout the design process, we demonstrate the ability to flexibly experiment with optimizing design transformations, and alternative multicore target platforms to achieve efficient implementations that are tailored to the resource constraints of these platforms.",
author = "Renjie Xie and Heikki Huttunen and Shuoxin Lin and Shuvra Bhattacharyya and Jarmo Takala",
note = "INT=tie,{"}Xie, Renjie{"}",
year = "2016",
month = "9",
language = "English",
publisher = "IEEE",
pages = "1862--1866",
booktitle = "2016 24th European Signal Processing Conference (EUSIPCO)",

}

RIS (suitable for import to EndNote) - Lataa

TY - GEN

T1 - Resource-Constrained Implementation and Optimization of a Deep Neural Network for Vehicle Classification

AU - Xie, Renjie

AU - Huttunen, Heikki

AU - Lin, Shuoxin

AU - Bhattacharyya, Shuvra

AU - Takala, Jarmo

N1 - INT=tie,"Xie, Renjie"

PY - 2016/9

Y1 - 2016/9

N2 - Deep learning has attracted great research interest in recent years in many signal processing application areas. However, investigation of deep learning implementations in highly resource-constrained contexts has been relatively unexplored due to the large computational requirements involved. In this paper, we investigate the implementation of a deep learning application for vehicle classification on multicore platforms with limited numbers of available processor cores. We apply model-based design methods based on signal processing oriented dataflow models of computation, and using the resulting dataflow representations, we apply various design optimizations to derive efficient implementations on three different multicore platforms. Using model-based design techniques throughout the design process, we demonstrate the ability to flexibly experiment with optimizing design transformations, and alternative multicore target platforms to achieve efficient implementations that are tailored to the resource constraints of these platforms.

AB - Deep learning has attracted great research interest in recent years in many signal processing application areas. However, investigation of deep learning implementations in highly resource-constrained contexts has been relatively unexplored due to the large computational requirements involved. In this paper, we investigate the implementation of a deep learning application for vehicle classification on multicore platforms with limited numbers of available processor cores. We apply model-based design methods based on signal processing oriented dataflow models of computation, and using the resulting dataflow representations, we apply various design optimizations to derive efficient implementations on three different multicore platforms. Using model-based design techniques throughout the design process, we demonstrate the ability to flexibly experiment with optimizing design transformations, and alternative multicore target platforms to achieve efficient implementations that are tailored to the resource constraints of these platforms.

M3 - Conference contribution

SP - 1862

EP - 1866

BT - 2016 24th European Signal Processing Conference (EUSIPCO)

PB - IEEE

ER -