基于残差量化卷积神经网络的人脸识别方法
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家重点研发计划(2017YFC0803700);上海市科委项目(17511101702);临港地区智能制造产业专项(#ZN2016020103);复旦大学工程与应用技术研究院先导项目(gyy2017-003)


Face Recognition Method Based on Quantized Residual Convolutional Neural Networks
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 增强出版
  • |
  • 文章评论
    摘要:

    针对大规模人脸识别问题,基于残差学习的超深卷积神经网络模型能取得比其他方法更高的识别精度,然而模型中存在的海量浮点参数需要占用大量的计算和存储资源,无法满足资源受限的场合需求.针对这一问题,本文设计了一种基于网络参数量化的超深残差网络模型.具体在Face-ResNet模型的基础上,增加了批归一化层和dropout层,加深了网络层次,对网络模型参数进行了二值量化,在模型识别精度损失极小的情况下,大幅压缩了模型大小并提升了计算效率.通过理论分析与实验验证了本文设计方法的有效性.

    Abstract:

    Very deep convolutional neural networks based on residual learning have achieved higher accuracy than other methods for large scale face recognition problem. But the massive floating-point parameters existing in the models need to occupy extensive computational and memory resources, which cannot be satisfied with the demand of occasions with limited resources. Aimed at the solution of this issue, a very deep residual neural network based on network model parameters quantization was designed in this study. In detail, based on the model Face-ResNet, the network was added with batch normalization layers and dropout layers, and also its total layers were deepened. Applying binary quantization to parameters of the designed network models, it can compress the model size substantially and improve computational efficiency with little loss of model recognition accuracy. Both theoretical analysis and experiments prove the effectiveness of the designed method.

    参考文献
    相似文献
    引证文献
引用本文

周光朕,杜姗姗,冯瑞,欧丽君,刘斌.基于残差量化卷积神经网络的人脸识别方法.计算机系统应用,2018,27(8):35-41

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2018-01-02
  • 最后修改日期:2018-01-23
  • 录用日期:
  • 在线发布日期: 2018-08-04
  • 出版日期:
您是第位访问者
版权所有:中国科学院软件研究所 京ICP备05046678号-3
地址:北京海淀区中关村南四街4号 中科院软件园区 7号楼305房间,邮政编码:100190
电话:010-62661041 传真: Email:csa (a) iscas.ac.cn
技术支持:北京勤云科技发展有限公司

京公网安备 11040202500063号