Publication detail

Implementation of the "Local Rank Differences" Image Feature Using SIMD Instructions of CPU

HEROUT, A. ZEMČÍK, P. JURÁNEK, R. HRADIŠ, M.

Original Title

Implementation of the "Local Rank Differences" Image Feature Using SIMD Instructions of CPU

English Title

Implementation of the "Local Rank Differences" Image Feature Using SIMD Instructions of CPU

Type

conference paper

Language

en

Original Abstract

Usage of statistical classifiers, namely AdaBoost and its modifications, in object detection and pattern recognition is a contemporary and popular trend . The computatiponal performance of these classifiers largely depends on low level image features they are using: both from the point of view of the amount of information the feature provides and the executional time of its evaluation. Local Rank Difference is an image feature that is alternative to commonly used Haar features. It is suitable for implementation in programmable (FPGA) or specialized (ASIC) hardware as well as graphics hardware (GPU). Additionally, as shown in this paper, it performs very well on common CPU. The paper discusses the LRD features and their properties, describes an experimental implementation of LRD using the multimedia instruction set of current general-purpose processors, presents its empirical performance measures compared to alternative approaches, and suggests several notes on practical usage of LRD and proposes directions for future work.

English abstract

Usage of statistical classifiers, namely AdaBoost and its modifications, in object detection and pattern recognition is a contemporary and popular trend . The computatiponal performance of these classifiers largely depends on low level image features they are using: both from the point of view of the amount of information the feature provides and the executional time of its evaluation. Local Rank Difference is an image feature that is alternative to commonly used Haar features. It is suitable for implementation in programmable (FPGA) or specialized (ASIC) hardware as well as graphics hardware (GPU). Additionally, as shown in this paper, it performs very well on common CPU. The paper discusses the LRD features and their properties, describes an experimental implementation of LRD using the multimedia instruction set of current general-purpose processors, presents its empirical performance measures compared to alternative approaches, and suggests several notes on practical usage of LRD and proposes directions for future work.

Keywords

Local Rank Differences, LRD, AdaBoost, Object Detection

RIV year

2008

Released

15.10.2008

Publisher

IEEE Computer Society

Location

Bhubaneswar

ISBN

978-0-7695-3476-3

Book

Proceedings of Sixth Indian Conference on Computer Vision, Graphics and Image Processing

Edition

NEUVEDEN

Edition number

NEUVEDEN

Pages from

1

Pages to

9

Pages count

9

URL

Documents

BibTex


@inproceedings{BUT32119,
  author="Adam {Herout} and Pavel {Zemčík} and Roman {Juránek} and Michal {Hradiš}",
  title="Implementation of the "Local Rank Differences" Image Feature Using SIMD Instructions of CPU",
  annote="Usage of statistical classifiers, namely AdaBoost and its modifications, in
object detection and pattern recognition is a contemporary and popular trend .
The computatiponal performance of these classifiers largely depends on low level
image features they are using: both from the point of view of the amount of
information the feature provides and the executional time of its evaluation.
Local Rank Difference is an image feature that is alternative to commonly used
Haar features. It is suitable for implementation in programmable (FPGA) or
specialized (ASIC) hardware as well as graphics hardware (GPU). Additionally, as
shown in this paper, it performs very well on common CPU. The paper discusses the
LRD features and their properties, describes an experimental implementation of
LRD using the multimedia instruction set of current general-purpose processors,
presents its empirical performance measures compared to alternative approaches,
and suggests several notes on practical usage of LRD and proposes directions for
future work.",
  address="IEEE Computer Society",
  booktitle="Proceedings of Sixth Indian Conference on Computer Vision, Graphics and Image Processing",
  chapter="32119",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Computer Society",
  year="2008",
  month="october",
  pages="1--9",
  publisher="IEEE Computer Society",
  type="conference paper"
}