An efficient method of constructing L1-type norm feature to estimate euclidean distance for fast vector quantization

Zhibin Pan, Koji Kotani, Tadahiro Ohmi

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

Abstract

In order to speed up the search process of vector quantization (VQ), it is most important to avoid computing k-dimensional Euclidean distance as many as possible. In order to find a best-matched codeword (winner) in the codebook for a certain input vector, it is a general way to roughly estimate other than exactly compute Euclidean distance immediately for the purpose of rejecting a candidate codeword. The lower dimensional features of a vector such as sum or the mean (L1 norm) and L2 norm are widely used for this purpose. Obviously, how to construct a suitable feature is a core problem for estimating Euclidean distance. In this paper, an efficient method of constructing L1-type norm feature is proposed by introducing a reference vector. In addition, the criterion on how to select an optimal reference vector is also given. Experimental results confirmed the effectiveness of the proposed method.

Original languageEnglish
Title of host publicationImage Processing, Biomedicine, Multimedia, Financial Engineering and Manufacturing - International Forum on Multimedia Image Processing, IFMIP - Proceedings of the Sixth Biannual World Automation Cong
EditorsM. Jamshidi, Y. Hata, A. Kamrani, J.S. Jamshidi
Pages217-222
Number of pages6
Publication statusPublished - 2004 Dec 1
EventImage Processing, Biomedicine, Multimedia, Financial Engineering and Manufacturing - International Forum on Multimedia Image Processing, IFMIP - Sixth Biannual World Automation Congress, WAC 2004 - Seville, Spain
Duration: 2004 Jun 262004 Jul 1

Publication series

NameImage Processing, Biomedicine, Multimedia, Financial Engineering and Manufacturing - Proceedings of the Sixth Biannual World Automation Congress

Other

OtherImage Processing, Biomedicine, Multimedia, Financial Engineering and Manufacturing - International Forum on Multimedia Image Processing, IFMIP - Sixth Biannual World Automation Congress, WAC 2004
CountrySpain
CitySeville
Period04/6/2604/7/1

Keywords

  • Euclidean distance
  • Fast vector quantization
  • L-type norm feature

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'An efficient method of constructing L<sub>1</sub>-type norm feature to estimate euclidean distance for fast vector quantization'. Together they form a unique fingerprint.

Cite this