Convolutional neural network implementations using Vitis AI

Akihiko Ushiroyama, Minoru Watanabe, Nobuya Watanabe, Akira Nagoya

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Recently, Xilinx has provided a field programmable gate array (FPGA)-based Vitis AI development environment, which is a deep learning framework to accel-erate AI operations and to seek a suitable neural network construction for a target application. We have implemented convolutional neural networks of three types onto the Vitis AI development environment and then we have evaluated their performance, power consumption, design man-hours, and so on. Results confirmed the Vitis AI benefits. Most notably, the FPGA platform power consumption is 4.96 times less than that of a GPU.

Original languageEnglish
Title of host publication2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022
EditorsRajashree Paul
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages365-371
Number of pages7
ISBN (Electronic)9781665483032
DOIs
Publication statusPublished - 2022
Event12th IEEE Annual Computing and Communication Workshop and Conference, CCWC 2022 - Virtual, Las Vegas, United States
Duration: Jan 26 2022Jan 29 2022

Publication series

Name2022 IEEE 12th Annual Computing and Communication Workshop and Conference, CCWC 2022

Conference

Conference12th IEEE Annual Computing and Communication Workshop and Conference, CCWC 2022
Country/TerritoryUnited States
CityVirtual, Las Vegas
Period1/26/221/29/22

Keywords

  • Convolutional Neural Network (CNN)
  • FPGA
  • Vitis AI

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Convolutional neural network implementations using Vitis AI'. Together they form a unique fingerprint.

Cite this