Interest point detection using imbalance oriented selection

Qi Li, Jieping Ye, Chandra Kambhamettu

Research output: Contribution to journalArticle

26 Citations (Scopus)

Abstract

Interest point detection has a wide range of applications, such as image retrieval and object recognition. Given an image, many previous interest point detectors first assign interest strength to each image point using a certain filtering technique, and then apply non-maximum suppression scheme to select a set of interest point candidates. However, we observe that non-maximum suppression tends to over-suppress good candidates for a weakly textured image such as a face image. We propose a new candidate selection scheme that chooses image points whose zero-/first-order intensities can be clustered into two imbalanced classes (in size), as candidates. Our tests of repeatability across image rotations and lighting conditions show the advantage of imbalance oriented selection. We further present a new face recognition application-facial identity representability evaluation-to show the value of imbalance oriented selection.

Original languageEnglish (US)
Pages (from-to)672-688
Number of pages17
JournalPattern Recognition
Volume41
Issue number2
DOIs
StatePublished - Feb 2008

Fingerprint

Object recognition
Image retrieval
Face recognition
Lighting
Detectors

Keywords

  • Facial expression
  • Interest point detection
  • Repeatability

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Signal Processing
  • Electrical and Electronic Engineering

Cite this

Interest point detection using imbalance oriented selection. / Li, Qi; Ye, Jieping; Kambhamettu, Chandra.

In: Pattern Recognition, Vol. 41, No. 2, 02.2008, p. 672-688.

Research output: Contribution to journalArticle

Li, Qi ; Ye, Jieping ; Kambhamettu, Chandra. / Interest point detection using imbalance oriented selection. In: Pattern Recognition. 2008 ; Vol. 41, No. 2. pp. 672-688.
@article{d2e43339117c43c3846d602df9745100,
title = "Interest point detection using imbalance oriented selection",
abstract = "Interest point detection has a wide range of applications, such as image retrieval and object recognition. Given an image, many previous interest point detectors first assign interest strength to each image point using a certain filtering technique, and then apply non-maximum suppression scheme to select a set of interest point candidates. However, we observe that non-maximum suppression tends to over-suppress good candidates for a weakly textured image such as a face image. We propose a new candidate selection scheme that chooses image points whose zero-/first-order intensities can be clustered into two imbalanced classes (in size), as candidates. Our tests of repeatability across image rotations and lighting conditions show the advantage of imbalance oriented selection. We further present a new face recognition application-facial identity representability evaluation-to show the value of imbalance oriented selection.",
keywords = "Facial expression, Interest point detection, Repeatability",
author = "Qi Li and Jieping Ye and Chandra Kambhamettu",
year = "2008",
month = "2",
doi = "10.1016/j.patcog.2007.06.020",
language = "English (US)",
volume = "41",
pages = "672--688",
journal = "Pattern Recognition",
issn = "0031-3203",
publisher = "Elsevier Limited",
number = "2",

}

TY - JOUR

T1 - Interest point detection using imbalance oriented selection

AU - Li, Qi

AU - Ye, Jieping

AU - Kambhamettu, Chandra

PY - 2008/2

Y1 - 2008/2

N2 - Interest point detection has a wide range of applications, such as image retrieval and object recognition. Given an image, many previous interest point detectors first assign interest strength to each image point using a certain filtering technique, and then apply non-maximum suppression scheme to select a set of interest point candidates. However, we observe that non-maximum suppression tends to over-suppress good candidates for a weakly textured image such as a face image. We propose a new candidate selection scheme that chooses image points whose zero-/first-order intensities can be clustered into two imbalanced classes (in size), as candidates. Our tests of repeatability across image rotations and lighting conditions show the advantage of imbalance oriented selection. We further present a new face recognition application-facial identity representability evaluation-to show the value of imbalance oriented selection.

AB - Interest point detection has a wide range of applications, such as image retrieval and object recognition. Given an image, many previous interest point detectors first assign interest strength to each image point using a certain filtering technique, and then apply non-maximum suppression scheme to select a set of interest point candidates. However, we observe that non-maximum suppression tends to over-suppress good candidates for a weakly textured image such as a face image. We propose a new candidate selection scheme that chooses image points whose zero-/first-order intensities can be clustered into two imbalanced classes (in size), as candidates. Our tests of repeatability across image rotations and lighting conditions show the advantage of imbalance oriented selection. We further present a new face recognition application-facial identity representability evaluation-to show the value of imbalance oriented selection.

KW - Facial expression

KW - Interest point detection

KW - Repeatability

UR - http://www.scopus.com/inward/record.url?scp=34848854498&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34848854498&partnerID=8YFLogxK

U2 - 10.1016/j.patcog.2007.06.020

DO - 10.1016/j.patcog.2007.06.020

M3 - Article

AN - SCOPUS:34848854498

VL - 41

SP - 672

EP - 688

JO - Pattern Recognition

JF - Pattern Recognition

SN - 0031-3203

IS - 2

ER -