Optimization of Sliding-DCT Based Gaussian Filtering for Hardware Accelerator

Tomoki Otsuka, Norishige Fukushima, Yoshihiro Maeda, Kenjiro Sugimoto, Sei Ichiro Kamata

研究成果: Conference contribution

抄録

Gaussian filtering is a smoothing filter used in various tasks. The main disadvantage is the dependence of the processing time on its kernel radius. One solution is using a sliding-discreet cosine transform (DCT), a constant-time algorithm for the kernel radius, and it provides the best performance in terms of both speed and accuracy. However, the speed and accuracy differ according to the type of DCT used. We can also accelerate the sliding-DCT based Gaussian filter by hardware accelerators, but the acceleration requires modification of the algorithms. In this paper, we focus on the fused multiply-add (FMA) instruction of hardware accelerators in modern computer architectures. The FMA instruction simultaneously performs multiplication and addition, i.e.,ax+b. We proposed an acceleration method of the sliding-DCT based Gaussian filtering for the FMA instruction. Moreover, we evaluate the performance of it in terms of computational time and approximation accuracy.

本文言語English
ホスト出版物のタイトル2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020
出版社Institute of Electrical and Electronics Engineers Inc.
ページ423-426
ページ数4
ISBN(電子版)9781728180670
DOI
出版ステータスPublished - 2020 12 1
イベント2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020 - Virtual, Macau, China
継続期間: 2020 12 12020 12 4

出版物シリーズ

名前2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020

Conference

Conference2020 IEEE International Conference on Visual Communications and Image Processing, VCIP 2020
国/地域China
CityVirtual, Macau
Period20/12/120/12/4

ASJC Scopus subject areas

  • コンピュータ ネットワークおよび通信
  • 信号処理
  • メディア記述

フィンガープリント

「Optimization of Sliding-DCT Based Gaussian Filtering for Hardware Accelerator」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル