Spatial interest pixels (SIPs): Useful low-level features of visual media data

Qi Li, Jieping Ye, Chandra Kambhamettu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Visual media data such as an image is the raw data representation for many important applications. Reducing the dimensionality of raw visual media data is desirable since high dimensionality degrades not only the effectiveness but also the efficiency of visual recognition algorithms. We present a comparative study on spatial interest pixels (SIPs), including eight-way (a novel SIP detector), Harris, and Lucas-Kanade, whose extraction is considered as an important step in reducing the dimensionality of visual media data. With extensive case studies, we have shown the usefulness of SIPs as low-level features of visual media data. A class-preserving dimension reduction algorithm (using GSVD) is applied to further reduce the dimension of feature vectors based on SIPs. The experiments showed its superiority over PCA.

Original languageEnglish (US)
Pages (from-to)89-108
Number of pages20
JournalMultimedia Tools and Applications
Volume30
Issue number1
DOIs
StatePublished - Jul 2006

Fingerprint

Pixel
Pixels
Dimensionality
Recognition Algorithm
Dimension Reduction
Feature Vector
Comparative Study
Detectors
Detector
Vision
Experiments
Experiment

Keywords

  • Dimensionreduction
  • Face recognition
  • Facial expression recognition
  • Low-levelfeatures
  • Spatial interest pixels

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Information Systems
  • Computer Graphics and Computer-Aided Design
  • Software
  • Theoretical Computer Science
  • Computational Theory and Mathematics

Cite this

Spatial interest pixels (SIPs) : Useful low-level features of visual media data. / Li, Qi; Ye, Jieping; Kambhamettu, Chandra.

In: Multimedia Tools and Applications, Vol. 30, No. 1, 07.2006, p. 89-108.

Research output: Contribution to journalArticle

Li, Qi ; Ye, Jieping ; Kambhamettu, Chandra. / Spatial interest pixels (SIPs) : Useful low-level features of visual media data. In: Multimedia Tools and Applications. 2006 ; Vol. 30, No. 1. pp. 89-108.
@article{b86f472a100e477e811c4d0945604d30,
title = "Spatial interest pixels (SIPs): Useful low-level features of visual media data",
abstract = "Visual media data such as an image is the raw data representation for many important applications. Reducing the dimensionality of raw visual media data is desirable since high dimensionality degrades not only the effectiveness but also the efficiency of visual recognition algorithms. We present a comparative study on spatial interest pixels (SIPs), including eight-way (a novel SIP detector), Harris, and Lucas-Kanade, whose extraction is considered as an important step in reducing the dimensionality of visual media data. With extensive case studies, we have shown the usefulness of SIPs as low-level features of visual media data. A class-preserving dimension reduction algorithm (using GSVD) is applied to further reduce the dimension of feature vectors based on SIPs. The experiments showed its superiority over PCA.",
keywords = "Dimensionreduction, Face recognition, Facial expression recognition, Low-levelfeatures, Spatial interest pixels",
author = "Qi Li and Jieping Ye and Chandra Kambhamettu",
year = "2006",
month = "7",
doi = "10.1007/s11042-006-0009-3",
language = "English (US)",
volume = "30",
pages = "89--108",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Spatial interest pixels (SIPs)

T2 - Useful low-level features of visual media data

AU - Li, Qi

AU - Ye, Jieping

AU - Kambhamettu, Chandra

PY - 2006/7

Y1 - 2006/7

N2 - Visual media data such as an image is the raw data representation for many important applications. Reducing the dimensionality of raw visual media data is desirable since high dimensionality degrades not only the effectiveness but also the efficiency of visual recognition algorithms. We present a comparative study on spatial interest pixels (SIPs), including eight-way (a novel SIP detector), Harris, and Lucas-Kanade, whose extraction is considered as an important step in reducing the dimensionality of visual media data. With extensive case studies, we have shown the usefulness of SIPs as low-level features of visual media data. A class-preserving dimension reduction algorithm (using GSVD) is applied to further reduce the dimension of feature vectors based on SIPs. The experiments showed its superiority over PCA.

AB - Visual media data such as an image is the raw data representation for many important applications. Reducing the dimensionality of raw visual media data is desirable since high dimensionality degrades not only the effectiveness but also the efficiency of visual recognition algorithms. We present a comparative study on spatial interest pixels (SIPs), including eight-way (a novel SIP detector), Harris, and Lucas-Kanade, whose extraction is considered as an important step in reducing the dimensionality of visual media data. With extensive case studies, we have shown the usefulness of SIPs as low-level features of visual media data. A class-preserving dimension reduction algorithm (using GSVD) is applied to further reduce the dimension of feature vectors based on SIPs. The experiments showed its superiority over PCA.

KW - Dimensionreduction

KW - Face recognition

KW - Facial expression recognition

KW - Low-levelfeatures

KW - Spatial interest pixels

UR - http://www.scopus.com/inward/record.url?scp=33748506097&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33748506097&partnerID=8YFLogxK

U2 - 10.1007/s11042-006-0009-3

DO - 10.1007/s11042-006-0009-3

M3 - Article

AN - SCOPUS:33748506097

VL - 30

SP - 89

EP - 108

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

IS - 1

ER -