Saliency Detection and Object Classification

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

Abstract

Humans have a distinct ability to process only the information that is of interest within a scene, however, this is not an easy task for computers. Trying to replicate this behaviour, many methods have been proposed to generate saliency maps that segment the object of interest within an image. In this paper, we investigate the problem of object classification, and whether saliency detection can be used. We generate saliency maps produced by two different currently published saliency detection methods, and train separate linear SVMs using the feature vectors obtained from these methods. We evaluate these methods against the traditional approach of extracting features from an image for object classification, namely HoG features. Our results show that saliency detection can be used for object classification, and improves accuracy by 5%.
LanguageEnglish
Title of host publicationUnknown Host Publication
Number of pages7
Publication statusAccepted/In press - 30 Jun 2017
EventIrish Machine Vision and Image Processing - Maynooth
Duration: 30 Jun 2017 → …

Conference

ConferenceIrish Machine Vision and Image Processing
Period30/06/17 → …

Keywords

  • Image Processing
  • Saliency Detection
  • Classification

Cite this

Cooley, C., Coleman, SA., Gardiner, B., & Bryan, S. (Accepted/In press). Saliency Detection and Object Classification. In Unknown Host Publication
@inproceedings{92f45f71dcca48d28ed19652eb67415c,
title = "Saliency Detection and Object Classification",
abstract = "Humans have a distinct ability to process only the information that is of interest within a scene, however, this is not an easy task for computers. Trying to replicate this behaviour, many methods have been proposed to generate saliency maps that segment the object of interest within an image. In this paper, we investigate the problem of object classification, and whether saliency detection can be used. We generate saliency maps produced by two different currently published saliency detection methods, and train separate linear SVMs using the feature vectors obtained from these methods. We evaluate these methods against the traditional approach of extracting features from an image for object classification, namely HoG features. Our results show that saliency detection can be used for object classification, and improves accuracy by 5{\%}.",
keywords = "Image Processing, Saliency Detection, Classification",
author = "Christopher Cooley and SA Coleman and Bryan Gardiner and Scotney Bryan",
year = "2017",
month = "6",
day = "30",
language = "English",
booktitle = "Unknown Host Publication",

}

Cooley, C, Coleman, SA, Gardiner, B & Bryan, S 2017, Saliency Detection and Object Classification. in Unknown Host Publication. Irish Machine Vision and Image Processing, 30/06/17.

Saliency Detection and Object Classification. / Cooley, Christopher; Coleman, SA; Gardiner, Bryan; Bryan, Scotney.

Unknown Host Publication. 2017.

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

TY - GEN

T1 - Saliency Detection and Object Classification

AU - Cooley, Christopher

AU - Coleman, SA

AU - Gardiner, Bryan

AU - Bryan, Scotney

PY - 2017/6/30

Y1 - 2017/6/30

N2 - Humans have a distinct ability to process only the information that is of interest within a scene, however, this is not an easy task for computers. Trying to replicate this behaviour, many methods have been proposed to generate saliency maps that segment the object of interest within an image. In this paper, we investigate the problem of object classification, and whether saliency detection can be used. We generate saliency maps produced by two different currently published saliency detection methods, and train separate linear SVMs using the feature vectors obtained from these methods. We evaluate these methods against the traditional approach of extracting features from an image for object classification, namely HoG features. Our results show that saliency detection can be used for object classification, and improves accuracy by 5%.

AB - Humans have a distinct ability to process only the information that is of interest within a scene, however, this is not an easy task for computers. Trying to replicate this behaviour, many methods have been proposed to generate saliency maps that segment the object of interest within an image. In this paper, we investigate the problem of object classification, and whether saliency detection can be used. We generate saliency maps produced by two different currently published saliency detection methods, and train separate linear SVMs using the feature vectors obtained from these methods. We evaluate these methods against the traditional approach of extracting features from an image for object classification, namely HoG features. Our results show that saliency detection can be used for object classification, and improves accuracy by 5%.

KW - Image Processing

KW - Saliency Detection

KW - Classification

M3 - Conference contribution

BT - Unknown Host Publication

ER -