An Enhanced Hybrid MobileNet

Hong Yen Chen, Chung Yen Su*

*此作品的通信作者

研究成果: 書貢獻/報告類型會議論文篇章

70 引文 斯高帕斯(Scopus)

摘要

Complicated and deep neural network models can achieve high accuracy for image recognition. However, they require a huge amount of computations and model parameters, which are not suitable for mobile and embedded devices. Therefore, MobileNet was proposed, which can reduce the number of parameters and computational cost dramatically. The main idea of MobileNet is to use a depthwise separable convolution. Two hyper-parameters, a width multiplier and a resolution multiplier are used to the trade-off between the accuracy and the latency. In this paper, we propose a new architecture to improve the MobileNet. Instead of using the resolution multiplier, we use a depth multiplier and combine with either Fractional Max Pooling or the max pooling. Experimental results on CIFAR database show that the proposed architecture can reduce the amount of computational cost and increase the accuracy simultaneously.

原文英語
主出版物標題2018 9th International Conference on Awareness Science and Technology, iCAST 2018
發行者Institute of Electrical and Electronics Engineers Inc.
頁面308-312
頁數5
ISBN(電子)9781538658260
DOIs
出版狀態已發佈 - 2018 10月 31
事件9th International Conference on Awareness Science and Technology, iCAST 2018 - Fukuoka, 日本
持續時間: 2018 9月 192018 9月 21

出版系列

名字2018 9th International Conference on Awareness Science and Technology, iCAST 2018

會議

會議9th International Conference on Awareness Science and Technology, iCAST 2018
國家/地區日本
城市Fukuoka
期間2018/09/192018/09/21

ASJC Scopus subject areas

  • 電腦網路與通信
  • 人機介面
  • 資訊系統與管理
  • 實驗與認知心理學
  • 社會心理學
  • 通訊

指紋

深入研究「An Enhanced Hybrid MobileNet」主題。共同形成了獨特的指紋。

引用此